------------------------------------------------------------------------------------------------------------------------------- name: log: c:\acdbookrevision\stata_final_programs_2013\racd09.txt log type: text opened on: 20 Jan 2013, 17:53:00 . . ********** OVERVIEW OF racd09.do ********** . . * STATA Program . * copyright C 2013 by A. Colin Cameron and Pravin K. Trivedi . * used for "Regression Analyis of Count Data" SECOND EDITION . * by A. Colin Cameron and Pravin K. Trivedi (2013) . * Cambridge University Press . . * This STATA program analyzes patent data for chapter 9 . * 9.4 FIXED EFFECTS and POOLED (PA / GEE) ESTIMATORS . * 9.5 RANDOM EFFECTS ESTIMATORS . * 9.8 DYNAMIC MODELS WITH RANDOM and FIXED EFFECTS . * It provides much more detail than is in the book . . * It takes a long time due to bootstraps to get panel robust se's . * In a couple of places bootstrap and jackknifes are commented out . * to speed up the program. . * In those cases the output is included as a comment. . . * To run you need file . * racd09data.dta . * in your directory . . ********** SETUP ********** . . set more off . version 12 . clear all . * set linesize 82 . set scheme s1mono /* Graphics scheme */ . * set maxvar 100 width 1000 . . ********** DATA DESCRIPTION . . * The original data is from . * Bronwyn Hall, Zvi Griliches, and Jerry Hausman (1986), . * "Patents and R&D: Is There a Lag?", . * International Economic Review, 27, 265-283. . * See this article for more detailed discussion . * Also see racd09makedata.do for further details . . ********** 9.4 PATENTS: READ DATA AND SUMMARIZE . . * This program gets clustered standard errors by bootstrap or jackknife . * when these are not provided by XT commands . * To speed up the program . * 1. Drop vce(jackkife) where it appears (this has been done here) . * 2. Reduce BREPS from 400 . . global BREPS 400 . . use racd09data.dta, clear . . *** TABLE 9.1: FREQUENCY DISTRIBUTION and SUMMARY STATISTICS for patents . . generate PATRANGE = PAT . recode PATRANGE (0=0) (1/5=1) (6/10=6) (11/20=11) (21/50=21) (51/100=51) (101/200=101) (201/515=201) (PATRANGE: 1124 changes made) . tabulate PATRANGE PATRANGE | Freq. Percent Cum. ------------+----------------------------------- 0 | 337 19.48 19.48 1 | 565 32.66 52.14 6 | 139 8.03 60.17 11 | 140 8.09 68.27 21 | 223 12.89 81.16 51 | 146 8.44 89.60 101 | 107 6.18 95.78 201 | 73 4.22 100.00 ------------+----------------------------------- Total | 1,730 100.00 . summarize PAT, detail Number of (successful) patents applied for this year ------------------------------------------------------------- Percentiles Smallest 1% 0 0 5% 0 0 10% 0 0 Obs 1730 25% 1 0 Sum of Wgt. 1730 50% 5 Mean 34.77168 Largest Std. Dev. 70.87538 75% 33 487 90% 105 495 Variance 5023.319 95% 182 508 Skewness 3.455604 99% 378 515 Kurtosis 17.11843 . . * Variable descriptions and summary statistics . describe Contains data from racd09data.dta obs: 1,730 vars: 24 7 Jun 2011 11:01 size: 166,080 ------------------------------------------------------------------------------------------------------------------------------- storage display value variable name type format label variable label ------------------------------------------------------------------------------------------------------------------------------- OBSNO float %9.0g YEAR float %9.0g Year CUSIP float %9.0g Compustat identifier for the firm ARDSIC float %9.0g Two-digit code for the applied R&D industrial classification SCISECT float %9.0g Equals 1 if firm in the scientific sector LOGK float %9.0g Logarithm of the book value of capital in 1972 SUMPAT float %9.0g Sum of patents applied for between 1972-1979 PAT float %9.0g Number of (successful) patents applied for this year PAT1 float %9.0g Number of (successful) patents applied for lagged one year PAT2 float %9.0g Number of (successful) patents applied for lagged two years PAT3 float %9.0g Number of (successful) patents applied for lagged three years PAT4 float %9.0g Number of (successful) patents applied for lagged four years LOGR float %9.0g Logarithm of R&D spending this year (in 1972$) LOGR1 float %9.0g Logarithm of R&D spending lagged one year (in 1972$) LOGR2 float %9.0g Logarithm of R&D spending lagged two years (in 1972$) LOGR3 float %9.0g Logarithm of R&D spending lagged three years (in 1972$) LOGR4 float %9.0g Logarithm of R&D spending lagged four years (in 1972$) LOGR5 float %9.0g Logarithm of R&D spending lagged five years (in 1972$) id float %9.0g id dyear2 float %9.0g = 1 if YEAR = 2 dyear3 float %9.0g = 1 if YEAR = 3 dyear4 float %9.0g = 1 if YEAR = 4 dyear5 float %9.0g = 1 if YEAR = 5 PATRANGE float %9.0g ------------------------------------------------------------------------------------------------------------------------------- Sorted by: id YEAR Note: dataset has changed since last saved . summarize Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- OBSNO | 1730 173.5 99.91006 1 346 YEAR | 1730 3 1.414622 1 5 CUSIP | 1730 531201.2 281748.4 800 989399 ARDSIC | 1730 -19.18497 169.1684 -999 21 SCISECT | 1730 .4248555 .494464 0 1 -------------+-------------------------------------------------------- LOGK | 1730 3.921063 2.093117 -1.76965 9.66626 SUMPAT | 1730 284.7312 570.4526 0 3806 PAT | 1730 34.77168 70.87538 0 515 PAT1 | 1730 35.87341 72.76243 0 528 PAT2 | 1730 36.70289 75.12335 0 595 -------------+-------------------------------------------------------- PAT3 | 1730 36.7185 75.52676 0 595 PAT4 | 1730 37.1711 76.53968 0 595 LOGR | 1730 1.256163 2.006314 -3.84868 7.03432 LOGR1 | 1730 1.233574 1.984091 -3.84868 7.06524 LOGR2 | 1730 1.218499 1.966808 -3.84868 7.06524 -------------+-------------------------------------------------------- LOGR3 | 1730 1.205683 1.951968 -3.84868 7.06524 LOGR4 | 1730 1.196941 1.942034 -3.67395 7.06524 LOGR5 | 1730 1.203451 1.934293 -3.67395 7.06524 id | 1730 173.5 99.91006 1 346 dyear2 | 1730 .2 .4001157 0 1 -------------+-------------------------------------------------------- dyear3 | 1730 .2 .4001157 0 1 dyear4 | 1730 .2 .4001157 0 1 dyear5 | 1730 .2 .4001157 0 1 PATRANGE | 1730 23.43815 45.57332 0 201 . . * Panel view . tsset panel variable: id (strongly balanced) time variable: YEAR, 1 to 5 delta: 1 unit . xtdescribe id: 1, 2, ..., 346 n = 346 YEAR: 1, 2, ..., 5 T = 5 Delta(YEAR) = 1 unit Span(YEAR) = 5 periods (id*YEAR uniquely identifies each observation) Distribution of T_i: min 5% 25% 50% 75% 95% max 5 5 5 5 5 5 5 Freq. Percent Cum. | Pattern ---------------------------+--------- 346 100.00 100.00 | 11111 ---------------------------+--------- 346 100.00 | XXXXX . xtsum PAT LOGR LOGR5 dyear2 dyear5 LOGK SCISECT Variable | Mean Std. Dev. Min Max | Observations -----------------+--------------------------------------------+---------------- PAT overall | 34.77168 70.87538 0 515 | N = 1730 between | 69.83142 0 473 | n = 346 within | 12.57671 -157.6283 200.3717 | T = 5 | | LOGR overall | 1.256163 2.006314 -3.84868 7.03432 | N = 1730 between | 1.996594 -3.415914 6.898732 | n = 346 within | .2193893 -.041151 2.388007 | T = 5 | | LOGR5 overall | 1.203451 1.934293 -3.67395 7.06524 | N = 1730 between | 1.917687 -2.99075 6.924144 | n = 346 within | .2692134 -.1899074 4.062701 | T = 5 | | dyear2 overall | .2 .4001157 0 1 | N = 1730 between | 0 .2 .2 | n = 346 within | .4001157 0 1 | T = 5 | | dyear5 overall | .2 .4001157 0 1 | N = 1730 between | 0 .2 .2 | n = 346 within | .4001157 0 1 | T = 5 | | LOGK overall | 3.921063 2.093117 -1.76965 9.66626 | N = 1730 between | 2.095542 -1.76965 9.66626 | n = 346 within | 0 3.921063 3.921063 | T = 5 | | SCISECT overall | .4248555 .494464 0 1 | N = 1730 between | .4950369 0 1 | n = 346 within | 0 .4248555 .4248555 | T = 5 . . * Serial correlation in PAT and LOGR . forvalues j = 1/4 { 2. quietly corr PAT L`j'.PAT 3. display "Autocorrelation at lag `j' = " %6.3f r(rho) 4. } Autocorrelation at lag 1 = 0.979 Autocorrelation at lag 2 = 0.966 Autocorrelation at lag 3 = 0.945 Autocorrelation at lag 4 = 0.946 . forvalues j = 1/4 { 2. quietly corr LOGR L`j'.LOGR 3. display "Autocorrelation at lag `j' = " %6.3f r(rho) 4. } Autocorrelation at lag 1 = 0.993 Autocorrelation at lag 2 = 0.987 Autocorrelation at lag 3 = 0.980 Autocorrelation at lag 4 = 0.973 . . * First-order autocorrelation differs in different year pairs . forvalues s = 2/4 { 2. quietly corr PAT L1.PAT if YEAR == `s' 3. display "Autocorrelation at lag 1 in year `s' = " %6.3f r(rho) 4. } Autocorrelation at lag 1 in year 2 = 0.991 Autocorrelation at lag 1 in year 3 = 0.985 Autocorrelation at lag 1 in year 4 = 0.970 . forvalues s = 2/4 { 2. quietly corr LOGR L1.LOGR if YEAR == `s' 3. display "Autocorrelation at lag 1 in year `s' = " %6.3f r(rho) 4. } Autocorrelation at lag 1 in year 2 = 0.993 Autocorrelation at lag 1 in year 3 = 0.992 Autocorrelation at lag 1 in year 4 = 0.994 . . by id: egen PATMEAN = mean(PAT) . generate PATDEV = PAT - PATMEAN . sort YEAR . summarize PATMEAN in 1/346, detail PATMEAN ------------------------------------------------------------- Percentiles Smallest 1% 0 0 5% 0 0 10% .2 0 Obs 346 25% 1.4 0 Sum of Wgt. 346 50% 5 Mean 34.77168 Largest Std. Dev. 69.83142 75% 33 329.4 90% 105.8 421.2 Variance 4876.427 95% 181.6 433.4 Skewness 3.329953 99% 329.4 473 Kurtosis 15.79888 . bysort YEAR: sum PAT ------------------------------------------------------------------------------------------------------------------------------- -> YEAR = 1 Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- PAT | 346 36.87283 75.98788 0 508 ------------------------------------------------------------------------------------------------------------------------------- -> YEAR = 2 Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- PAT | 346 35.84682 73.31613 0 487 ------------------------------------------------------------------------------------------------------------------------------- -> YEAR = 3 Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- PAT | 346 36.23121 72.75146 0 456 ------------------------------------------------------------------------------------------------------------------------------- -> YEAR = 4 Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- PAT | 346 32.80636 65.6505 0 434 ------------------------------------------------------------------------------------------------------------------------------- -> YEAR = 5 Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- PAT | 346 32.10116 66.36197 0 515 . . sort id YEAR . . ********** ASIDE: CROSS SECTION MODELS . . global XLIST LOGR LOGR1 LOGR2 LOGR3 LOGR4 LOGR5 LOGK SCISECT dyear2 dyear3 dyear4 dyear5 . * global XLIST LOGR LOGR1 LOGR2 LOGR3 LOGR4 LOGR5 dyear2 dyear3 dyear4 dyear5 LOGK SCISECT . global XLISTTIMEVARYING LOGR LOGR1 LOGR2 LOGR3 LOGR4 LOGR5 dyear2 dyear3 dyear4 dyear5 . . * Serial correlation in residuals . quietly poisson PAT $XLIST . predict poissrawresid, score . forvalues j = 1/4 { 2. quietly corr poissrawresid L`j'.poissrawresid 3. display "Autocorrelation at lag `j' = " %6.3f r(rho) 4. } Autocorrelation at lag 1 = 0.946 Autocorrelation at lag 2 = 0.902 Autocorrelation at lag 3 = 0.851 Autocorrelation at lag 4 = 0.847 . forvalues s = 2/4 { 2. quietly corr poissrawresid L1.poissrawresid if YEAR == `s' 3. display "Autocorrelation at lag 1 in year `s' = " %6.3f r(rho) 4. } Autocorrelation at lag 1 in year 2 = 0.973 Autocorrelation at lag 1 in year 3 = 0.961 Autocorrelation at lag 1 in year 4 = 0.927 . . * Serial correlation in residuals . quietly poisson PAT $XLIST i.id . predict poissrawresid2, score . forvalues j = 1/4 { 2. quietly corr poissrawresid2 L`j'.poissrawresid2 3. display "Autocorrelation at lag `j' = " %6.3f r(rho) 4. } Autocorrelation at lag 1 = 0.234 Autocorrelation at lag 2 = -0.374 Autocorrelation at lag 3 = -0.741 Autocorrelation at lag 4 = -0.554 . forvalues s = 2/4 { 2. quietly corr poissrawresid2 L1.poissrawresid2 if YEAR == `s' 3. display "Autocorrelation at lag 1 in year `s' = " %6.3f r(rho) 4. } Autocorrelation at lag 1 in year 2 = 0.640 Autocorrelation at lag 1 in year 3 = -0.018 Autocorrelation at lag 1 in year 4 = -0.076 . . * Poisson Cross-section . quietly poisson PAT $XLIST . estimates store PCSdef . quietly poisson PAT $XLIST, vce(robust) . estimates store PCSrob . * Following standard errors are preferred . quietly poisson PAT $XLIST, vce(cluster id) . estimates store PCSclu . display "Table 9.2: first column Sum ln R" Table 9.2: first column Sum ln R . lincom LOGR + LOGR1 + LOGR2 + LOGR3 + LOGR4 + LOGR5 ( 1) [PAT]LOGR + [PAT]LOGR1 + [PAT]LOGR2 + [PAT]LOGR3 + [PAT]LOGR4 + [PAT]LOGR5 = 0 ------------------------------------------------------------------------------ PAT | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- (1) | .4856263 .0752866 6.45 0.000 .3380673 .6331854 ------------------------------------------------------------------------------ . . * Negative binomial NB2 Cross-section . quietly nbreg PAT $XLIST . estimates store NBCSdef . quietly nbreg PAT $XLIST, vce(robust) . estimates store NBCSrob . * Following standard errors are preferred . quietly nbreg PAT $XLIST, vce(cluster id) . lincom LOGR + LOGR1 + LOGR2 + LOGR3 + LOGR4 + LOGR5 ( 1) [PAT]LOGR + [PAT]LOGR1 + [PAT]LOGR2 + [PAT]LOGR3 + [PAT]LOGR4 + [PAT]LOGR5 = 0 ------------------------------------------------------------------------------ PAT | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- (1) | .8100443 .0636399 12.73 0.000 .6853124 .9347762 ------------------------------------------------------------------------------ . estimates store NBCSclu . . estimates table PCSdef PCSrob PCSclu NBCSdef NBCSrob NBCSclu, b(%7.4f) se(%7.3f) stats(N ll) -------------------------------------------------------------------------- Variable | PCSdef PCSrob PCSclu NBCSdef NBCSrob NBCSclu -------------+------------------------------------------------------------ PAT | LOGR | 0.1345 0.1345 0.1345 0.4311 0.4311 0.4311 | 0.031 0.180 0.183 0.112 0.141 0.133 LOGR1 | -0.0529 -0.0529 -0.0529 -0.1171 -0.1171 -0.1171 | 0.043 0.242 0.106 0.156 0.186 0.141 LOGR2 | 0.0082 0.0082 0.0082 0.1065 0.1065 0.1065 | 0.040 0.232 0.093 0.150 0.168 0.121 LOGR3 | 0.0661 0.0661 0.0661 0.0764 0.0764 0.0764 | 0.037 0.221 0.114 0.137 0.155 0.103 LOGR4 | 0.0902 0.0902 0.0902 0.1938 0.1938 0.1938 | 0.033 0.198 0.093 0.125 0.128 0.088 LOGR5 | 0.2395 0.2395 0.2395 0.1194 0.1194 0.1194 | 0.022 0.132 0.123 0.085 0.090 0.086 LOGK | 0.2529 0.2529 0.2529 0.1013 0.1013 0.1013 | 0.004 0.028 0.059 0.024 0.027 0.054 SCISECT | 0.4543 0.4543 0.4543 -0.0046 -0.0046 -0.0046 | 0.009 0.077 0.167 0.056 0.059 0.119 dyear2 | -0.0435 -0.0435 -0.0435 -0.0558 -0.0558 -0.0558 | 0.013 0.096 0.018 0.077 0.076 0.035 dyear3 | -0.0524 -0.0524 -0.0524 -0.0609 -0.0609 -0.0609 | 0.013 0.097 0.030 0.077 0.080 0.043 dyear4 | -0.1702 -0.1702 -0.1702 -0.1220 -0.1220 -0.1220 | 0.014 0.094 0.046 0.077 0.085 0.047 dyear5 | -0.2019 -0.2019 -0.2019 -0.2267 -0.2267 -0.2267 | 0.014 0.089 0.046 0.077 0.085 0.049 _cons | 0.8099 0.8099 0.8099 0.9088 0.9088 0.9088 | 0.021 0.130 0.242 0.097 0.105 0.182 -------------+------------------------------------------------------------ lnalpha | _cons | -0.2660 -0.2660 -0.2660 | 0.044 0.048 0.089 -------------+------------------------------------------------------------ Statistics | N | 1730 1730 1730 1730 1730 1730 ll | -1.8e+04 -1.8e+04 -1.8e+04 -5.8e+03 -5.8e+03 -5.8e+03 -------------------------------------------------------------------------- legend: b/se . . ********* 9.4 POPULATION AVERAGED MODELS . . * Poisson GEE exchangeable . quietly xtgee PAT $XLIST, family(poisson) corr(exch) . estimates store PPAEXdef . * Following standard errors are preferred . xtgee PAT $XLIST, family(poisson) corr(exch) vce(robust) Iteration 1: tolerance = .1414367 Iteration 2: tolerance = .02055835 Iteration 3: tolerance = .00182509 Iteration 4: tolerance = .00017476 Iteration 5: tolerance = .00001718 Iteration 6: tolerance = 1.732e-06 Iteration 7: tolerance = 1.776e-07 GEE population-averaged model Number of obs = 1730 Group variable: id Number of groups = 346 Link: log Obs per group: min = 5 Family: Poisson avg = 5.0 Correlation: exchangeable max = 5 Wald chi2(12) = 671.74 Scale parameter: 1 Prob > chi2 = 0.0000 (Std. Err. adjusted for clustering on id) ------------------------------------------------------------------------------ | Robust PAT | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- LOGR | .3155815 .061931 5.10 0.000 .194199 .4369639 LOGR1 | -.0522286 .0602 -0.87 0.386 -.1702185 .0657612 LOGR2 | .104817 .0535308 1.96 0.050 -.0001014 .2097354 LOGR3 | .0196527 .0674715 0.29 0.771 -.1125891 .1518945 LOGR4 | .0229611 .0535738 0.43 0.668 -.0820415 .1279638 LOGR5 | .0488891 .0551695 0.89 0.376 -.0592411 .1570192 LOGK | .2698966 .0566612 4.76 0.000 .1588426 .3809505 SCISECT | .4402067 .1751261 2.51 0.012 .0969658 .7834476 dyear2 | -.0455811 .0170808 -2.67 0.008 -.0790589 -.0121034 dyear3 | -.0462482 .0259597 -1.78 0.075 -.0971283 .0046319 dyear4 | -.1685656 .0408254 -4.13 0.000 -.248582 -.0885492 dyear5 | -.2135843 .0413192 -5.17 0.000 -.2945685 -.1326002 _cons | .7774219 .2448108 3.18 0.001 .2976016 1.257242 ------------------------------------------------------------------------------ . estimates store PPAEXrob . display "Table 9.2: second column Sum ln R" Table 9.2: second column Sum ln R . lincom LOGR + LOGR1 + LOGR2 + LOGR3 + LOGR4 + LOGR5 ( 1) LOGR + LOGR1 + LOGR2 + LOGR3 + LOGR4 + LOGR5 = 0 ------------------------------------------------------------------------------ PAT | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- (1) | .4596728 .0698585 6.58 0.000 .3227527 .5965929 ------------------------------------------------------------------------------ . * Poisson GEE AR(1) errors . quietly xtgee PAT $XLIST, family(poisson) corr(ar1) vce(robust) . estimates store PPAARrob . . * Negative binomial . * Here we need to give the value of alpha - use the cross-section nbreg estimate . * Furthermore needs to be done manually (can't use alpha where scalar alpha = e(alpha)) . quietly nbreg PAT $XLIST . scalar alpha = e(alpha) . display "The correct alpha to use is: " alpha The correct alpha to use is: .76642446 . * NB2 GEE exchangeable errors errors . xtgee PAT $XLIST, family(nbinomial .7785956) corr(exch) Iteration 1: tolerance = .08658775 Iteration 2: tolerance = .01420478 Iteration 3: tolerance = .00226872 Iteration 4: tolerance = .00040058 Iteration 5: tolerance = .00008206 Iteration 6: tolerance = .00001543 Iteration 7: tolerance = 2.957e-06 Iteration 8: tolerance = 5.594e-07 GEE population-averaged model Number of obs = 1730 Group variable: id Number of groups = 346 Link: log Obs per group: min = 5 Family: negative binomial(k=1.2844) avg = 5.0 Correlation: exchangeable max = 5 Wald chi2(12) = 1275.79 Scale parameter: 1 Prob > chi2 = 0.0000 ------------------------------------------------------------------------------ PAT | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- LOGR | .5204098 .0791055 6.58 0.000 .3653659 .6754537 LOGR1 | -.0848338 .0933313 -0.91 0.363 -.2677599 .0980922 LOGR2 | .1249174 .086829 1.44 0.150 -.0452643 .2950991 LOGR3 | .0533976 .0805725 0.66 0.508 -.1045216 .2113169 LOGR4 | .0914135 .0737452 1.24 0.215 -.0531243 .2359514 LOGR5 | .0236248 .057445 0.41 0.681 -.0889652 .1362149 LOGK | .163459 .0409921 3.99 0.000 .0831159 .243802 SCISECT | .0687045 .1045165 0.66 0.511 -.1361441 .273553 dyear2 | -.0535706 .0445095 -1.20 0.229 -.1408075 .0336664 dyear3 | -.0575273 .0449518 -1.28 0.201 -.1456311 .0305765 dyear4 | -.123724 .0450283 -2.75 0.006 -.2119779 -.0354701 dyear5 | -.2392881 .0453723 -5.27 0.000 -.3282162 -.1503599 _cons | .7333156 .1558079 4.71 0.000 .4279378 1.038693 ------------------------------------------------------------------------------ . estimates store NBPAEXdef . * Following standard errors are preferred . xtgee PAT $XLIST, family(nbinomial .7785956) corr(exch) vce(robust) Iteration 1: tolerance = .08658775 Iteration 2: tolerance = .01420478 Iteration 3: tolerance = .00226872 Iteration 4: tolerance = .00040058 Iteration 5: tolerance = .00008206 Iteration 6: tolerance = .00001543 Iteration 7: tolerance = 2.957e-06 Iteration 8: tolerance = 5.594e-07 GEE population-averaged model Number of obs = 1730 Group variable: id Number of groups = 346 Link: log Obs per group: min = 5 Family: negative binomial(k=1.2844) avg = 5.0 Correlation: exchangeable max = 5 Wald chi2(12) = 1707.70 Scale parameter: 1 Prob > chi2 = 0.0000 (Std. Err. adjusted for clustering on id) ------------------------------------------------------------------------------ | Semirobust PAT | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- LOGR | .5204098 .107983 4.82 0.000 .308767 .7320526 LOGR1 | -.0848338 .112758 -0.75 0.452 -.3058354 .1361678 LOGR2 | .1249174 .0852358 1.47 0.143 -.0421418 .2919765 LOGR3 | .0533976 .0989001 0.54 0.589 -.140443 .2472383 LOGR4 | .0914135 .0826048 1.11 0.268 -.0704888 .2533159 LOGR5 | .0236248 .0614235 0.38 0.701 -.096763 .1440127 LOGK | .163459 .0466632 3.50 0.000 .0720007 .2549172 SCISECT | .0687045 .1137579 0.60 0.546 -.1542568 .2916658 dyear2 | -.0535706 .0337001 -1.59 0.112 -.1196216 .0124805 dyear3 | -.0575273 .0403097 -1.43 0.154 -.1365329 .0214782 dyear4 | -.123724 .0460129 -2.69 0.007 -.2139076 -.0335404 dyear5 | -.2392881 .0486633 -4.92 0.000 -.3346663 -.1439098 _cons | .7333156 .1686302 4.35 0.000 .4028064 1.063825 ------------------------------------------------------------------------------ . estimates store NBPAEXrob . * NB2 GEE AR(1) errors . xtgee PAT $XLIST, family(nbinomial .7785956) corr(ar1) vce(robust) Iteration 1: tolerance = .09358647 Iteration 2: tolerance = .00925672 Iteration 3: tolerance = .00101627 Iteration 4: tolerance = .00020977 Iteration 5: tolerance = .00002707 Iteration 6: tolerance = 6.827e-06 Iteration 7: tolerance = 1.108e-06 Iteration 8: tolerance = 2.525e-07 GEE population-averaged model Number of obs = 1730 Group and time vars: id YEAR Number of groups = 346 Link: log Obs per group: min = 5 Family: negative binomial(k=1.2844) avg = 5.0 Correlation: AR(1) max = 5 Wald chi2(12) = 1635.21 Scale parameter: 1 Prob > chi2 = 0.0000 (Std. Err. adjusted for clustering on id) ------------------------------------------------------------------------------ | Semirobust PAT | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- LOGR | .4628448 .1057626 4.38 0.000 .2555538 .6701357 LOGR1 | -.057346 .1045565 -0.55 0.583 -.2622729 .1475809 LOGR2 | .1465882 .0958547 1.53 0.126 -.0412836 .33446 LOGR3 | .0759155 .1030232 0.74 0.461 -.1260062 .2778373 LOGR4 | .1447976 .0742072 1.95 0.051 -.0006458 .290241 LOGR5 | .014246 .0643692 0.22 0.825 -.1119153 .1404074 LOGK | .1191441 .0487287 2.45 0.014 .0236376 .2146507 SCISECT | .0114246 .1151492 0.10 0.921 -.2142637 .237113 dyear2 | -.0534989 .0342032 -1.56 0.118 -.1205358 .0135381 dyear3 | -.0546555 .0423031 -1.29 0.196 -.137568 .0282571 dyear4 | -.1148204 .047687 -2.41 0.016 -.2082851 -.0213556 dyear5 | -.2301842 .0516966 -4.45 0.000 -.3315076 -.1288607 _cons | .8552731 .1749549 4.89 0.000 .5123678 1.198178 ------------------------------------------------------------------------------ . estimates store NBPAARrob . . estimates table PPAEXdef PPAEXrob PPAARrob NBPAEXdef NBPAEXrob NBPAARrob, b(%7.4f) se(%7.3f) stats(N ll) -------------------------------------------------------------------------- Variable | PPAEX~f PPAEX~b PPAAR~b NBPAE~f NBPAE~b NBPAA~b -------------+------------------------------------------------------------ LOGR | 0.3156 0.3156 0.2573 0.5204 0.5204 0.4628 | 0.014 0.062 0.056 0.079 0.108 0.106 LOGR1 | -0.0522 -0.0522 -0.0280 -0.0848 -0.0848 -0.0573 | 0.016 0.060 0.058 0.093 0.113 0.105 LOGR2 | 0.1048 0.1048 0.1201 0.1249 0.1249 0.1466 | 0.014 0.054 0.058 0.087 0.085 0.096 LOGR3 | 0.0197 0.0197 0.0459 0.0534 0.0534 0.0759 | 0.013 0.067 0.060 0.081 0.099 0.103 LOGR4 | 0.0230 0.0230 0.0537 0.0914 0.0914 0.1448 | 0.012 0.054 0.050 0.074 0.083 0.074 LOGR5 | 0.0489 0.0489 0.0268 0.0236 0.0236 0.0142 | 0.010 0.055 0.046 0.057 0.061 0.064 LOGK | 0.2699 0.2699 0.2623 0.1635 0.1635 0.1191 | 0.008 0.057 0.054 0.041 0.047 0.049 SCISECT | 0.4402 0.4402 0.4708 0.0687 0.0687 0.0114 | 0.019 0.175 0.170 0.105 0.114 0.115 dyear2 | -0.0456 -0.0456 -0.0453 -0.0536 -0.0536 -0.0535 | 0.005 0.017 0.017 0.045 0.034 0.034 dyear3 | -0.0462 -0.0462 -0.0447 -0.0575 -0.0575 -0.0547 | 0.005 0.026 0.026 0.045 0.040 0.042 dyear4 | -0.1686 -0.1686 -0.1633 -0.1237 -0.1237 -0.1148 | 0.005 0.041 0.042 0.045 0.046 0.048 dyear5 | -0.2136 -0.2136 -0.2053 -0.2393 -0.2393 -0.2302 | 0.005 0.041 0.041 0.045 0.049 0.052 _cons | 0.7774 0.7774 0.7498 0.7333 0.7333 0.8553 | 0.039 0.245 0.234 0.156 0.169 0.175 -------------+------------------------------------------------------------ N | 1730 1730 1730 1730 1730 1730 ll | -------------------------------------------------------------------------- legend: b/se . . ********** 9.4 FIXED EFFECTS MODELS . . * Poisson fixed effects . quietly xtpoisson PAT $XLIST, fe . estimates store PFEdef . * Following standard errors are preferred . xtpoisson PAT $XLIST, fe vce(robust) note: 22 groups (110 obs) dropped because of all zero outcomes note: LOGK dropped because it is constant within group note: SCISECT dropped because it is constant within group Iteration 0: log pseudolikelihood = -3660.2656 Iteration 1: log pseudolikelihood = -3536.3518 Iteration 2: log pseudolikelihood = -3536.3086 Iteration 3: log pseudolikelihood = -3536.3086 Conditional fixed-effects Poisson regression Number of obs = 1620 Group variable: id Number of groups = 324 Obs per group: min = 5 avg = 5.0 max = 5 Wald chi2(10) = 48.17 Log pseudolikelihood = -3536.3086 Prob > chi2 = 0.0000 (Std. Err. adjusted for clustering on id) ------------------------------------------------------------------------------ | Robust PAT | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- LOGR | .3222105 .0807547 3.99 0.000 .1639341 .4804868 LOGR1 | -.0871295 .0712049 -1.22 0.221 -.2266885 .0524295 LOGR2 | .0785816 .0620597 1.27 0.205 -.0430532 .2002164 LOGR3 | .00106 .078183 0.01 0.989 -.1521758 .1542958 LOGR4 | -.0046414 .063583 -0.07 0.942 -.1292617 .119979 LOGR5 | .0026068 .0759235 0.03 0.973 -.1462004 .1514141 dyear2 | -.0426076 .0167407 -2.55 0.011 -.0754187 -.0097965 dyear3 | -.0400462 .0248168 -1.61 0.107 -.0886862 .0085939 dyear4 | -.1571185 .035894 -4.38 0.000 -.2274694 -.0867676 dyear5 | -.1980306 .0368759 -5.37 0.000 -.2703059 -.1257552 ------------------------------------------------------------------------------ . estimates store PFErob . display "Table 9.2: third column Sum ln R" Table 9.2: third column Sum ln R . lincom LOGR + LOGR1 + LOGR2 + LOGR3 + LOGR4 + LOGR5 ( 1) [PAT]LOGR + [PAT]LOGR1 + [PAT]LOGR2 + [PAT]LOGR3 + [PAT]LOGR4 + [PAT]LOGR5 = 0 ------------------------------------------------------------------------------ PAT | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- (1) | .312688 .1431872 2.18 0.029 .0320462 .5933298 ------------------------------------------------------------------------------ . . * Following checks with a jackknife . * xtpoisson PAT $XLIST, fe vce(jackknife) . * estimates store PFEjack . * Dummy variables - gives same estimates and almost same standard errors . * poisson PAT $XLIST i.id, vce(cluster id) . * estimates store PFEDVclu . . * Negative binomial fixed effects . quietly xtnbreg PAT $XLIST, fe . estimates store NBFEdef . * Following takes a while - may want to comment out . * xtnbreg PAT $XLIST, fe vce(jacknife) // Jackknife as no vce(robust) here . * estimates store NBFEjack . display "Table 9.2: fourth column Sum ln R" Table 9.2: fourth column Sum ln R . lincom LOGR + LOGR1 + LOGR2 + LOGR3 + LOGR4 + LOGR5 ( 1) [PAT]LOGR + [PAT]LOGR1 + [PAT]LOGR2 + [PAT]LOGR3 + [PAT]LOGR4 + [PAT]LOGR5 = 0 ------------------------------------------------------------------------------ PAT | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- (1) | .1929697 .0756175 2.55 0.011 .0447621 .3411774 ------------------------------------------------------------------------------ . * Dummy variables - gives different estimates . * nbreg PAT $XLIST i.id, vce(cluster id) dispersion(constant) . * estimates store NBFEDVclu . . estimates table PFEdef PFErob NBFEdef, b(%7.4f) se(%7.3f) stats(N ll) equations(1) -------------------------------------------- Variable | PFEdef PFErob NBFEdef -------------+------------------------------ LOGR | 0.3222 0.3222 0.2727 | 0.046 0.081 0.071 LOGR1 | -0.0871 -0.0871 -0.0979 | 0.049 0.071 0.077 LOGR2 | 0.0786 0.0786 0.0321 | 0.045 0.062 0.071 LOGR3 | 0.0011 0.0011 -0.0204 | 0.041 0.078 0.066 LOGR4 | -0.0046 -0.0046 0.0162 | 0.038 0.064 0.063 LOGR5 | 0.0026 0.0026 -0.0097 | 0.032 0.076 0.053 dyear2 | -0.0426 -0.0426 -0.0384 | 0.013 0.017 0.024 dyear3 | -0.0400 -0.0400 -0.0399 | 0.013 0.025 0.025 dyear4 | -0.1571 -0.1571 -0.1443 | 0.014 0.036 0.026 dyear5 | -0.1980 -0.1980 -0.1958 | 0.015 0.037 0.027 LOGK | 0.2071 | 0.078 SCISECT | 0.0176 | 0.198 _cons | 1.6614 | 0.344 -------------+------------------------------ N | 1620 1620 1620 ll | -3.5e+03 -3.5e+03 -3.2e+03 -------------------------------------------- legend: b/se . . /* . xtnbreg PAT $XLIST, fe vce(jacknife) // Jackknife as no vce(robust) here > (running xtnbreg on estimation sample) > > Jackknife replications (324) > ----+--- 1 ---+--- 2 ---+--- 3 ---+--- 4 ---+--- 5 > .................................................. 50 > .................................................. 100 > .................................................. 150 > .................................................. 200 > .................................................. 250 > .................................................. 300 > ........................ > Conditional FE negative binomial regression Number of obs = 1620 > Group variable: id Number of groups = 324 > > Obs per group: min = 5 > avg = 5.0 > max = 5 > > F( 12, 323) = 6.96 > Log likelihood = -3203.0644 Prob > F = 0.0000 > > (Replications based on 324 clusters in id) > ------------------------------------------------------------------------------ > | Jackknife > PAT | Coef. Std. Err. t P>|t| [95% Conf. Interval] > -------------+---------------------------------------------------------------- > LOGR | .272679 .0801407 3.40 0.001 .1150154 .4303426 > LOGR1 | -.0978866 .0795833 -1.23 0.220 -.2544537 .0586804 > LOGR2 | .0320762 .0604324 0.53 0.596 -.0868146 .150967 > LOGR3 | -.0203923 .072151 -0.28 0.778 -.1623375 .121553 > LOGR4 | .0162214 .0625026 0.26 0.795 -.1067422 .139185 > LOGR5 | -.009728 .0663792 -0.15 0.884 -.1403181 .1208621 > LOGK | .2071488 .1029215 2.01 0.045 .0046676 .40963 > SCISECT | .0176397 .3264891 0.05 0.957 -.6246739 .6599533 > dyear2 | -.0383927 .0179525 -2.14 0.033 -.0737113 -.003074 > dyear3 | -.0399403 .0253008 -1.58 0.115 -.0897155 .0098349 > dyear4 | -.1443278 .0319978 -4.51 0.000 -.2072783 -.0813773 > dyear5 | -.1957518 .0339705 -5.76 0.000 -.2625832 -.1289204 > _cons | 1.661392 .5358439 3.10 0.002 .6072074 2.715577 > ------------------------------------------------------------------------------ > . estimates store NBFEjack > > . lincom LOGR + LOGR1 + LOGR2 + LOGR3 + LOGR4 + LOGR5 > ( 1) [PAT]LOGR + [PAT]LOGR1 + [PAT]LOGR2 + [PAT]LOGR3 + [PAT]LOGR4 + [PAT]LOGR5 = 0 > ------------------------------------------------------------------------------ > PAT | Coef. Std. Err. t P>|t| [95% Conf. Interval] > -------------+---------------------------------------------------------------- > (1) | .1929697 .1136149 1.70 0.090 -.0305488 .4164883 > ------------------------------------------------------------------------------ > > */ . . *** TABLE 9.2: POOLED POISSON, POOLED GEE, POISSON FE, NB1 FE . . * Note: Following gives default se's for NB1FE and not jackknife se's (given above) . estimates table PCSclu PPAEXrob PFErob NBFEdef, b(%11.4f) se(%11.3f) stats(N ll) equations(1) ---------------------------------------------------------------------- Variable | PCSclu PPAEXrob PFErob NBFEdef -------------+-------------------------------------------------------- LOGR | 0.1345 0.3156 0.3222 0.2727 | 0.183 0.062 0.081 0.071 LOGR1 | -0.0529 -0.0522 -0.0871 -0.0979 | 0.106 0.060 0.071 0.077 LOGR2 | 0.0082 0.1048 0.0786 0.0321 | 0.093 0.054 0.062 0.071 LOGR3 | 0.0661 0.0197 0.0011 -0.0204 | 0.114 0.067 0.078 0.066 LOGR4 | 0.0902 0.0230 -0.0046 0.0162 | 0.093 0.054 0.064 0.063 LOGR5 | 0.2395 0.0489 0.0026 -0.0097 | 0.123 0.055 0.076 0.053 LOGK | 0.2529 0.2699 0.2071 | 0.059 0.057 0.078 SCISECT | 0.4543 0.4402 0.0176 | 0.167 0.175 0.198 dyear2 | -0.0435 -0.0456 -0.0426 -0.0384 | 0.018 0.017 0.017 0.024 dyear3 | -0.0524 -0.0462 -0.0400 -0.0399 | 0.030 0.026 0.025 0.025 dyear4 | -0.1702 -0.1686 -0.1571 -0.1443 | 0.046 0.041 0.036 0.026 dyear5 | -0.2019 -0.2136 -0.1980 -0.1958 | 0.046 0.041 0.037 0.027 _cons | 0.8099 0.7774 1.6614 | 0.242 0.245 0.344 -------------+-------------------------------------------------------- N | 1730 1730 1620 1620 ll | -17834.1381 -3536.3086 -3203.0644 ---------------------------------------------------------------------- legend: b/se . . ********** 9.5 RANDOM EFFECTS MODELS . . * There is no robust option for , re so need to bootstrap . . * Poisson - gamma random effects . xtpoisson PAT $XLIST, re Fitting Poisson model: Iteration 0: log likelihood = -17836.658 Iteration 1: log likelihood = -17834.138 Iteration 2: log likelihood = -17834.138 Fitting full model: Iteration 0: log likelihood = -5303.2636 Iteration 1: log likelihood = -5241.765 Iteration 2: log likelihood = -5234.9526 Iteration 3: log likelihood = -5234.9265 Iteration 4: log likelihood = -5234.9265 Random-effects Poisson regression Number of obs = 1730 Group variable: id Number of groups = 346 Random effects u_i ~ Gamma Obs per group: min = 5 avg = 5.0 max = 5 Wald chi2(12) = 1272.14 Log likelihood = -5234.9265 Prob > chi2 = 0.0000 ------------------------------------------------------------------------------ PAT | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- LOGR | .4034537 .0435022 9.27 0.000 .318191 .4887165 LOGR1 | -.0461765 .0482224 -0.96 0.338 -.1406906 .0483376 LOGR2 | .1079235 .0447115 2.41 0.016 .0202905 .1955565 LOGR3 | .0297733 .0413235 0.72 0.471 -.0512193 .110766 LOGR4 | .0106957 .0377074 0.28 0.777 -.0632094 .0846008 LOGR5 | .0406111 .0315738 1.29 0.198 -.0212724 .1024946 LOGK | .2916932 .0393368 7.42 0.000 .2145945 .368792 SCISECT | .2570001 .1122716 2.29 0.022 .0369517 .4770484 dyear2 | -.0449624 .0131291 -3.42 0.001 -.070695 -.0192298 dyear3 | -.0483864 .0134018 -3.61 0.000 -.0746534 -.0221193 dyear4 | -.1741619 .0139702 -12.47 0.000 -.201543 -.1467809 dyear5 | -.2258977 .0146645 -15.40 0.000 -.2546396 -.1971557 _cons | .4107881 .1467443 2.80 0.005 .1231746 .6984016 -------------+---------------------------------------------------------------- /lnalpha | -.156739 .0809735 -.3154441 .0019661 -------------+---------------------------------------------------------------- alpha | .8549271 .0692264 .7294648 1.001968 ------------------------------------------------------------------------------ Likelihood-ratio test of alpha=0: chibar2(01) = 2.5e+04 Prob>=chibar2 = 0.000 . estimates store PREdef . * Following standard errors are preferred . xtpoisson PAT $XLIST, re vce(boot, reps($BREPS) seed(10101)) (running xtpoisson on estimation sample) Bootstrap replications (400) ----+--- 1 ---+--- 2 ---+--- 3 ---+--- 4 ---+--- 5 .................................................. 50 .................................................. 100 .................................................. 150 .................................................. 200 .................................................. 250 .................................................. 300 .................................................. 350 .................................................. 400 Random-effects Poisson regression Number of obs = 1730 Group variable: id Number of groups = 346 Random effects u_i ~ Gamma Obs per group: min = 5 avg = 5.0 max = 5 Wald chi2(12) = 1295.42 Log likelihood = -5234.9265 Prob > chi2 = 0.0000 (Replications based on 346 clusters in id) ------------------------------------------------------------------------------ | Observed Bootstrap Normal-based PAT | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- LOGR | .4034537 .0809342 4.98 0.000 .2448256 .5620819 LOGR1 | -.0461765 .0767807 -0.60 0.548 -.1966639 .1043109 LOGR2 | .1079235 .0644214 1.68 0.094 -.0183402 .2341872 LOGR3 | .0297733 .0840982 0.35 0.723 -.1350561 .1946028 LOGR4 | .0106957 .0673302 0.16 0.874 -.121269 .1426604 LOGR5 | .0406111 .0760807 0.53 0.593 -.1085044 .1897266 LOGK | .2916932 .0768476 3.80 0.000 .1410747 .4423117 SCISECT | .2570001 .1357081 1.89 0.058 -.008983 .5229831 dyear2 | -.0449624 .0177426 -2.53 0.011 -.0797373 -.0101875 dyear3 | -.0483864 .0274466 -1.76 0.078 -.1021808 .0054081 dyear4 | -.1741619 .0388339 -4.48 0.000 -.250275 -.0980488 dyear5 | -.2258977 .0392428 -5.76 0.000 -.3028121 -.1489833 _cons | .4107881 .2262128 1.82 0.069 -.032581 .8541571 -------------+---------------------------------------------------------------- /lnalpha | -.156739 .0986191 -.3500288 .0365508 -------------+---------------------------------------------------------------- alpha | .8549271 .0843121 .7046678 1.037227 ------------------------------------------------------------------------------ Likelihood-ratio test of alpha=0: chibar2(01) = 2.5e+04 Prob>=chibar2 = 0.000 . display "Table 9.3: first column Sum ln R" Table 9.3: first column Sum ln R . lincom LOGR + LOGR1 + LOGR2 + LOGR3 + LOGR4 + LOGR5 ( 1) [PAT]LOGR + [PAT]LOGR1 + [PAT]LOGR2 + [PAT]LOGR3 + [PAT]LOGR4 + [PAT]LOGR5 = 0 ------------------------------------------------------------------------------ PAT | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- (1) | .5462808 .0936551 5.83 0.000 .3627202 .7298415 ------------------------------------------------------------------------------ . estimates store PRErob . . * Poisson - normal random effects . xtpoisson PAT $XLIST, re normal Fitting comparison Poisson model: Iteration 0: log likelihood = -17836.658 Iteration 1: log likelihood = -17834.138 Iteration 2: log likelihood = -17834.138 Fitting full model: tau = 0.0 log likelihood = -17834.138 tau = 0.1 log likelihood = -6619.5905 tau = 0.2 log likelihood = -6432.6737 tau = 0.3 log likelihood = -6612.791 Iteration 0: log likelihood = -5643.779 Iteration 1: log likelihood = -5273.0769 Iteration 2: log likelihood = -5246.4625 Iteration 3: log likelihood = -5245.0169 Iteration 4: log likelihood = -5245.0127 Iteration 5: log likelihood = -5245.0127 Random-effects Poisson regression Number of obs = 1730 Group variable: id Number of groups = 346 Random effects u_i ~ Gaussian Obs per group: min = 5 avg = 5.0 max = 5 Wald chi2(12) = 1206.64 Log likelihood = -5245.0127 Prob > chi2 = 0.0000 ------------------------------------------------------------------------------ PAT | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- LOGR | .4153326 .0435856 9.53 0.000 .3299063 .5007588 LOGR1 | -.0403106 .0482797 -0.83 0.404 -.1349371 .0543159 LOGR2 | .112102 .0447765 2.50 0.012 .0243416 .1998624 LOGR3 | .0348424 .0413625 0.84 0.400 -.0462266 .1159114 LOGR4 | .0126664 .0377112 0.34 0.737 -.0612462 .0865789 LOGR5 | .0471076 .0315904 1.49 0.136 -.0148084 .1090236 LOGK | .2917346 .0418194 6.98 0.000 .2097701 .3736991 SCISECT | .443509 .1236684 3.59 0.000 .2011233 .6858946 dyear2 | -.0453001 .0131302 -3.45 0.001 -.0710349 -.0195653 dyear3 | -.0496512 .0134031 -3.70 0.000 -.0759208 -.0233817 dyear4 | -.1766992 .0139747 -12.64 0.000 -.2040892 -.1493093 dyear5 | -.2301051 .0146847 -15.67 0.000 -.2588867 -.2013236 _cons | -.1513461 .1687456 -0.90 0.370 -.4820813 .1793892 -------------+---------------------------------------------------------------- /lnsig2u | -.0052583 .0954254 -0.06 0.956 -.1922887 .1817721 -------------+---------------------------------------------------------------- sigma_u | .9973743 .0475874 .9083329 1.095144 ------------------------------------------------------------------------------ Likelihood-ratio test of sigma_u=0: chibar2(01) = 2.5e+04 Pr>=chibar2 = 0.000 . estimates store PRENdef . lincom LOGR + LOGR1 + LOGR2 + LOGR3 + LOGR4 + LOGR5 ( 1) [PAT]LOGR + [PAT]LOGR1 + [PAT]LOGR2 + [PAT]LOGR3 + [PAT]LOGR4 + [PAT]LOGR5 = 0 ------------------------------------------------------------------------------ PAT | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- (1) | .5817403 .0409286 14.21 0.000 .5015218 .6619588 ------------------------------------------------------------------------------ . * Following standard errors are preferred . * This took a long time so is comment out . * xtpoisson PAT $XLIST, re normal vce(jackknife) . * estimates store PRENrob . * display "Table 9.3: second column Sum ln R" . * lincom LOGR + LOGR1 + LOGR2 + LOGR3 + LOGR4 + LOGR5 . * or .. . * Note that about 20% of bootstrap replications failed to estimate . * xtpoisson PAT $XLIST, re normal vce(boot, reps($BREPS) seed(10101)) . * lincom LOGR + LOGR1 + LOGR2 + LOGR3 + LOGR4 + LOGR5 . . /* JACKKNIFE > . xtpoisson PAT $XLIST, re normal vce(jackknife) > (running xtpoisson on estimation sample) > Jackknife replications (346) > ----+--- 1 ---+--- 2 ---+--- 3 ---+--- 4 ---+--- 5 > .................................................. 50 > .................................................. 100 > ...x.............................................. 150 > .................................................. 200 > .................................................. 250 > .................................................. 300 > .............................................. > Random-effects Poisson regression Number of obs = 1730 > Group variable: id Number of groups = 346 > Random effects u_i ~ Gaussian Obs per group: min = 5 > avg = 5.0 > max = 5 > F( 12, 344) = 94.88 > Log likelihood = -5245.0127 Prob > F = 0.0000 > (Replications based on 346 clusters in id) > ------------------------------------------------------------------------------ > | Jackknife > PAT | Coef. Std. Err. t P>|t| [95% Conf. Interval] > -------------+---------------------------------------------------------------- > LOGR | .4153326 .0781077 5.32 0.000 .2617038 .5689614 > LOGR1 | -.0403106 .0744648 -0.54 0.589 -.1867743 .1061531 > LOGR2 | .112102 .0649979 1.72 0.085 -.0157413 .2399453 > LOGR3 | .0348424 .0886451 0.39 0.695 -.1395122 .209197 > LOGR4 | .0126664 .0669306 0.19 0.850 -.1189783 .144311 > LOGR5 | .0471076 .0794252 0.59 0.553 -.1091126 .2033278 > LOGK | .2917346 .0834706 3.50 0.001 .1275577 .4559115 > SCISECT | .443509 .1527652 2.90 0.004 .1430375 .7439805 > dyear2 | -.0453001 .0180533 -2.51 0.013 -.0808088 -.0097914 > dyear3 | -.0496512 .0270002 -1.84 0.067 -.1027575 .003455 > dyear4 | -.1766992 .0398763 -4.43 0.000 -.2551313 -.0982671 > dyear5 | -.2301051 .0389316 -5.91 0.000 -.306679 -.1535312 > _cons | -.1513461 .2729969 -0.55 0.580 -.6882993 .3856072 > -------------+---------------------------------------------------------------- > /lnsig2u | -.0052583 .1107198 -0.05 0.962 -.2230313 .2125146 > -------------+---------------------------------------------------------------- > sigma_u | .9973743 .0552145 .8944774 1.112108 > ------------------------------------------------------------------------------ > Likelihood-ratio test of sigma_u=0: chibar2(01) = 2.5e+04 Pr>=chibar2 = 0.000 > . lincom LOGR + LOGR1 + LOGR2 + LOGR3 + LOGR4 + LOGR5 > ( 1) [PAT]LOGR + [PAT]LOGR1 + [PAT]LOGR2 + [PAT]LOGR3 + [PAT]LOGR4 + [PAT]LOGR5 = 0 > ------------------------------------------------------------------------------ > PAT | Coef. Std. Err. t P>|t| [95% Conf. Interval] > -------------+---------------------------------------------------------------- > (1) | .5817403 .0977223 5.95 0.000 .3895319 .7739487 > ------------------------------------------------------------------------------ > */ . . /* BOOTSTRAP FOR POISSON RE NORMAL FAILED MANY TIMES > . xtpoisson PAT $XLIST, re normal vce(boot, reps($BREPS) seed(10101)) > (running xtpoisson on estimation sample) > Bootstrap replications (400) > ----+--- 1 ---+--- 2 ---+--- 3 ---+--- 4 ---+--- 5 > x...x..x..xx..x.xx....x...........x.............x. 50 > ......x...x......x.x.xx.xx..x...x.....x....x.x.... 100 > .x...................x.....x...x..x....x..x....... 150 > ..............x........x...x..x..........x...x.... 200 > .........x.......x....x...x...xx..xxx.....xx....x. 250 > xx.x.........x.x.x........x............x........x. 300 > .x.....x....................x.x..........x.....x.x 350 > x.........x..x...x.....x........x.....x........... 400 > Random-effects Poisson regression Number of obs = 1730 > Group variable: id Number of groups = 346 > Random effects u_i ~ Gaussian Obs per group: min = 5 > avg = 5.0 > max = 5 > Wald chi2(12) = 1202.83 > Log likelihood = -5245.0127 Prob > chi2 = 0.0000 > (Replications based on 346 clusters in id) > ------------------------------------------------------------------------------ > | Observed Bootstrap Normal-based > PAT | Coef. Std. Err. z P>|z| [95% Conf. Interval] > -------------+---------------------------------------------------------------- > LOGR | .4153326 .0816933 5.08 0.000 .2552166 .5754485 > LOGR1 | -.0403106 .0779906 -0.52 0.605 -.1931694 .1125482 > LOGR2 | .112102 .0632054 1.77 0.076 -.0117782 .2359822 > LOGR3 | .0348424 .0842935 0.41 0.679 -.1303698 .2000546 > LOGR4 | .0126664 .066196 0.19 0.848 -.1170754 .1424082 > LOGR5 | .0471076 .0728772 0.65 0.518 -.095729 .1899442 > LOGK | .2917346 .0782631 3.73 0.000 .1383418 .4451274 > SCISECT | .443509 .1454438 3.05 0.002 .1584444 .7285735 > dyear2 | -.0453001 .0181748 -2.49 0.013 -.0809221 -.0096781 > dyear3 | -.0496512 .0274114 -1.81 0.070 -.1033767 .0040742 > dyear4 | -.1766992 .0383587 -4.61 0.000 -.251881 -.1015175 > dyear5 | -.2301051 .0379868 -6.06 0.000 -.3045579 -.1556523 > _cons | -.1513461 .2651781 -0.57 0.568 -.6710856 .3683935 > -------------+---------------------------------------------------------------- > /lnsig2u | -.0052583 .120223 -0.04 0.965 -.240891 .2303744 > -------------+---------------------------------------------------------------- > sigma_u | .9973743 .0599537 .8865254 1.122083 > ------------------------------------------------------------------------------ > Likelihood-ratio test of sigma_u=0: chibar2(01) = 2.5e+04 Pr>=chibar2 = 0.000 > . estimates store PRENrob > . display "Table 9.3: second column Sum ln R" > Table 9.3: second column Sum ln R > . lincom LOGR + LOGR1 + LOGR2 + LOGR3 + LOGR4 + LOGR5 > ( 1) [PAT]LOGR + [PAT]LOGR1 + [PAT]LOGR2 + [PAT]LOGR3 + [PAT]LOGR4 + [PAT]LOGR5 = 0 > ------------------------------------------------------------------------------ > PAT | Coef. Std. Err. z P>|z| [95% Conf. Interval] > -------------+---------------------------------------------------------------- > (1) | .5817403 .0909148 6.40 0.000 .4035506 .75993 > ------------------------------------------------------------------------------ > */ . . /* SO INSTEAD PANEL JACKKNIFE FOR POISSON RE NORMAL > THESE ARE THE STANDARD ERRORS IN TABLE 9.3 FOR POISSON RE NORMAL > . xtpoisson PAT $XLIST, re normal vce(jackknife) > (running xtpoisson on estimation sample) > Jackknife replications (346) > ----+--- 1 ---+--- 2 ---+--- 3 ---+--- 4 ---+--- 5 > .................................................. 50 > .................................................. 100 > ...x.............................................. 150 > .................................................. 200 > .................................................. 250 > .................................................. 300 > .............................................. > Random-effects Poisson regression Number of obs = 1730 > Group variable: id Number of groups = 346 > > Random effects u_i ~ Gaussian Obs per group: min = 5 > avg = 5.0 > max = 5 > F( 12, 344) = 94.88 > Log likelihood = -5245.0127 Prob > F = 0.0000 > (Replications based on 346 clusters in id) > ------------------------------------------------------------------------------ > | Jackknife > PAT | Coef. Std. Err. t P>|t| [95% Conf. Interval] > -------------+---------------------------------------------------------------- > LOGR | .4153326 .0781077 5.32 0.000 .2617038 .5689614 > LOGR1 | -.0403106 .0744648 -0.54 0.589 -.1867743 .1061531 > LOGR2 | .112102 .0649979 1.72 0.085 -.0157413 .2399453 > LOGR3 | .0348424 .0886451 0.39 0.695 -.1395122 .209197 > LOGR4 | .0126664 .0669306 0.19 0.850 -.1189783 .144311 > LOGR5 | .0471076 .0794252 0.59 0.553 -.1091126 .2033278 > LOGK | .2917346 .0834706 3.50 0.001 .1275577 .4559115 > SCISECT | .443509 .1527652 2.90 0.004 .1430375 .7439805 > dyear2 | -.0453001 .0180533 -2.51 0.013 -.0808088 -.0097914 > dyear3 | -.0496512 .0270002 -1.84 0.067 -.1027575 .003455 > dyear4 | -.1766992 .0398763 -4.43 0.000 -.2551313 -.0982671 > dyear5 | -.2301051 .0389316 -5.91 0.000 -.306679 -.1535312 > _cons | -.1513461 .2729969 -0.55 0.580 -.6882993 .3856072 > -------------+---------------------------------------------------------------- > /lnsig2u | -.0052583 .1107198 -0.05 0.962 -.2230313 .2125146 > -------------+---------------------------------------------------------------- > sigma_u | .9973743 .0552145 .8944774 1.112108 > ------------------------------------------------------------------------------ > Likelihood-ratio test of sigma_u=0: chibar2(01) = 2.5e+04 Pr>=chibar2 = 0.000 > > . lincom LOGR + LOGR1 + LOGR2 + LOGR3 + LOGR4 + LOGR5 > ( 1) [PAT]LOGR + [PAT]LOGR1 + [PAT]LOGR2 + [PAT]LOGR3 + [PAT]LOGR4 + [PAT]LOGR5 = 0 > > ------------------------------------------------------------------------------ > PAT | Coef. Std. Err. t P>|t| [95% Conf. Interval] > -------------+---------------------------------------------------------------- > (1) | .5817403 .0977223 5.95 0.000 .3895319 .7739487 > ------------------------------------------------------------------------------ > */ . . * Negative binomial - beta distributed ratio random effects . xtnbreg PAT $XLIST, re Fitting negative binomial (constant dispersion) model: Iteration 0: log likelihood = -17836.658 Iteration 1: log likelihood = -17834.138 Iteration 2: log likelihood = -17834.138 Iteration 0: log likelihood = -37163.276 Iteration 1: log likelihood = -17331.718 Iteration 2: log likelihood = -8376.139 (backed up) Iteration 3: log likelihood = -6999.1967 Iteration 4: log likelihood = -6948.0162 Iteration 5: log likelihood = -6948.0022 Iteration 6: log likelihood = -6948.0022 Iteration 0: log likelihood = -6948.0022 Iteration 1: log likelihood = -6484.3647 (not concave) Iteration 2: log likelihood = -6063.4801 Iteration 3: log likelihood = -5996.6042 Iteration 4: log likelihood = -5954.4866 Iteration 5: log likelihood = -5954.1073 Iteration 6: log likelihood = -5954.1071 Fitting full model: Iteration 0: log likelihood = -5074.487 Iteration 1: log likelihood = -4961.2657 Iteration 2: log likelihood = -4948.6428 Iteration 3: log likelihood = -4948.4945 Iteration 4: log likelihood = -4948.4944 Random-effects negative binomial regression Number of obs = 1730 Group variable: id Number of groups = 346 Random effects u_i ~ Beta Obs per group: min = 5 avg = 5.0 max = 5 Wald chi2(12) = 944.21 Log likelihood = -4948.4944 Prob > chi2 = 0.0000 ------------------------------------------------------------------------------ PAT | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- LOGR | .3503119 .0652818 5.37 0.000 .2223619 .4782619 LOGR1 | -.0030317 .0750916 -0.04 0.968 -.1502085 .1441452 LOGR2 | .1049876 .0688488 1.52 0.127 -.0299537 .2399289 LOGR3 | .0163523 .0636376 0.26 0.797 -.1083752 .1410797 LOGR4 | .0359425 .0587161 0.61 0.540 -.0791389 .1510239 LOGR5 | .0718323 .0482887 1.49 0.137 -.0228119 .1664764 LOGK | .161937 .0417874 3.88 0.000 .0800351 .2438388 SCISECT | .1176419 .1066164 1.10 0.270 -.0913224 .3266063 dyear2 | -.0436736 .0213435 -2.05 0.041 -.085506 -.0018411 dyear3 | -.0556597 .0218572 -2.55 0.011 -.098499 -.0128203 dyear4 | -.1831055 .0227183 -8.06 0.000 -.2276326 -.1385784 dyear5 | -.2300438 .0231525 -9.94 0.000 -.2754219 -.1846658 _cons | .8995618 .1681113 5.35 0.000 .5700698 1.229054 -------------+---------------------------------------------------------------- /ln_r | .9877591 .0961426 .7993231 1.176195 /ln_s | .7009608 .1079684 .4893467 .9125748 -------------+---------------------------------------------------------------- r | 2.68521 .2581631 2.224035 3.242015 s | 2.015688 .2176306 1.63125 2.490728 ------------------------------------------------------------------------------ Likelihood-ratio test vs. pooled: chibar2(01) = 2011.23 Prob>=chibar2 = 0.000 . estimates store NBREdef . * Following standard errors are preferred . xtnbreg PAT $XLIST, re vce(boot, reps($BREPS) seed(10101)) (running xtnbreg on estimation sample) Bootstrap replications (400) ----+--- 1 ---+--- 2 ---+--- 3 ---+--- 4 ---+--- 5 .................................................. 50 .................................................. 100 .................................................. 150 .................................................. 200 .................................................. 250 .................................................. 300 .................................................. 350 .................................................. 400 Random-effects negative binomial regression Number of obs = 1730 Group variable: id Number of groups = 346 Random effects u_i ~ Beta Obs per group: min = 5 avg = 5.0 max = 5 Wald chi2(12) = 479.91 Log likelihood = -4948.4944 Prob > chi2 = 0.0000 (Replications based on 346 clusters in id) ------------------------------------------------------------------------------ | Observed Bootstrap Normal-based PAT | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- LOGR | .3503119 .0715707 4.89 0.000 .2100359 .4905879 LOGR1 | -.0030317 .0722299 -0.04 0.967 -.1445996 .1385362 LOGR2 | .1049876 .0580511 1.81 0.071 -.0087904 .2187656 LOGR3 | .0163523 .0769153 0.21 0.832 -.134399 .1671035 LOGR4 | .0359425 .0590841 0.61 0.543 -.0798602 .1517451 LOGR5 | .0718323 .0607418 1.18 0.237 -.0472195 .1908841 LOGK | .161937 .0539557 3.00 0.003 .0561857 .2676883 SCISECT | .1176419 .1391107 0.85 0.398 -.1550101 .3902939 dyear2 | -.0436736 .0174441 -2.50 0.012 -.0778634 -.0094837 dyear3 | -.0556597 .0257636 -2.16 0.031 -.1061554 -.005164 dyear4 | -.1831055 .0355249 -5.15 0.000 -.2527331 -.1134779 dyear5 | -.2300438 .0364212 -6.32 0.000 -.3014281 -.1586596 _cons | .8995618 .2145383 4.19 0.000 .4790745 1.320049 -------------+---------------------------------------------------------------- /ln_r | .9877591 .1620609 .6701257 1.305393 /ln_s | .7009608 .1298822 .4463964 .9555251 -------------+---------------------------------------------------------------- r | 2.68521 .4351675 1.954483 3.689137 s | 2.015688 .261802 1.562671 2.600036 ------------------------------------------------------------------------------ Likelihood-ratio test vs. pooled: chibar2(01) = 2011.23 Prob>=chibar2 = 0.000 . estimates store NBRErob . display "Table 9.3: third column Sum ln R" Table 9.3: third column Sum ln R . lincom LOGR + LOGR1 + LOGR2 + LOGR3 + LOGR4 + LOGR5 ( 1) [PAT]LOGR + [PAT]LOGR1 + [PAT]LOGR2 + [PAT]LOGR3 + [PAT]LOGR4 + [PAT]LOGR5 = 0 ------------------------------------------------------------------------------ PAT | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- (1) | .5763948 .0703515 8.19 0.000 .4385084 .7142813 ------------------------------------------------------------------------------ . * There is no RE normal option for xtnbreg . . estimates table PREdef PRErob PRENdef NBREdef NBRErob, b(%11.4f) se(%11.3f) stats(N ll) ------------------------------------------------------------------------------------ Variable | PREdef PRErob PRENdef NBREdef NBRErob -------------+---------------------------------------------------------------------- PAT | LOGR | 0.4035 0.4035 0.4153 0.3503 0.3503 | 0.044 0.081 0.044 0.065 0.072 LOGR1 | -0.0462 -0.0462 -0.0403 -0.0030 -0.0030 | 0.048 0.077 0.048 0.075 0.072 LOGR2 | 0.1079 0.1079 0.1121 0.1050 0.1050 | 0.045 0.064 0.045 0.069 0.058 LOGR3 | 0.0298 0.0298 0.0348 0.0164 0.0164 | 0.041 0.084 0.041 0.064 0.077 LOGR4 | 0.0107 0.0107 0.0127 0.0359 0.0359 | 0.038 0.067 0.038 0.059 0.059 LOGR5 | 0.0406 0.0406 0.0471 0.0718 0.0718 | 0.032 0.076 0.032 0.048 0.061 LOGK | 0.2917 0.2917 0.2917 0.1619 0.1619 | 0.039 0.077 0.042 0.042 0.054 SCISECT | 0.2570 0.2570 0.4435 0.1176 0.1176 | 0.112 0.136 0.124 0.107 0.139 dyear2 | -0.0450 -0.0450 -0.0453 -0.0437 -0.0437 | 0.013 0.018 0.013 0.021 0.017 dyear3 | -0.0484 -0.0484 -0.0497 -0.0557 -0.0557 | 0.013 0.027 0.013 0.022 0.026 dyear4 | -0.1742 -0.1742 -0.1767 -0.1831 -0.1831 | 0.014 0.039 0.014 0.023 0.036 dyear5 | -0.2259 -0.2259 -0.2301 -0.2300 -0.2300 | 0.015 0.039 0.015 0.023 0.036 _cons | 0.4108 0.4108 -0.1513 0.8996 0.8996 | 0.147 0.226 0.169 0.168 0.215 -------------+---------------------------------------------------------------------- lnalpha | _cons | -0.1567 -0.1567 | 0.081 0.099 -------------+---------------------------------------------------------------------- lnsig2u | _cons | -0.0053 | 0.095 -------------+---------------------------------------------------------------------- ln_r | _cons | 0.9878 0.9878 | 0.096 0.162 -------------+---------------------------------------------------------------------- ln_s | _cons | 0.7010 0.7010 | 0.108 0.130 -------------+---------------------------------------------------------------------- Statistics | N | 1730 1730 1730 1730 1730 ll | -5234.9265 -5234.9265 -5245.0127 -4948.4944 -4948.4944 ------------------------------------------------------------------------------------ legend: b/se . . * Conditionally Correlated Random Effects . sort id . by id: egen LOGRMEAN = mean(LOGR) . by id: egen LOGR1MEAN = mean(LOGR1) . by id: egen LOGR2MEAN = mean(LOGR2) . by id: egen LOGR3MEAN = mean(LOGR3) . by id: egen LOGR4MEAN = mean(LOGR4) . by id: egen LOGR5MEAN = mean(LOGR5) . global MEANS LOGRMEAN LOGR1MEAN LOGR2MEAN LOGR3MEAN LOGR4MEAN LOGR5MEAN . . * Poisson CCRE . xtpoisson PAT $XLIST $MEANS, re Fitting Poisson model: Iteration 0: log likelihood = -17636.972 Iteration 1: log likelihood = -17631.366 Iteration 2: log likelihood = -17631.366 Fitting full model: Iteration 0: log likelihood = -5248.2202 Iteration 1: log likelihood = -5214.6644 Iteration 2: log likelihood = -5211.9092 Iteration 3: log likelihood = -5211.898 Iteration 4: log likelihood = -5211.898 Random-effects Poisson regression Number of obs = 1730 Group variable: id Number of groups = 346 Random effects u_i ~ Gamma Obs per group: min = 5 avg = 5.0 max = 5 Wald chi2(18) = 1367.27 Log likelihood = -5211.898 Prob > chi2 = 0.0000 ------------------------------------------------------------------------------ PAT | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- LOGR | .3216624 .0459496 7.00 0.000 .2316029 .4117219 LOGR1 | -.0871244 .0487055 -1.79 0.074 -.1825855 .0083367 LOGR2 | .0788715 .044789 1.76 0.078 -.0089133 .1666564 LOGR3 | .0004448 .0414296 0.01 0.991 -.0807558 .0816454 LOGR4 | -.0047836 .0378581 -0.13 0.899 -.078984 .0694168 LOGR5 | .0024365 .0322758 0.08 0.940 -.060823 .0656959 LOGK | .0617355 .0496748 1.24 0.214 -.0356252 .1590963 SCISECT | -.0489817 .1190747 -0.41 0.681 -.2823639 .1844004 dyear2 | -.0425778 .0131311 -3.24 0.001 -.0683142 -.0168414 dyear3 | -.0399678 .0134663 -2.97 0.003 -.0663612 -.0135744 dyear4 | -.1570141 .0142265 -11.04 0.000 -.1848974 -.1291307 dyear5 | -.1978545 .0152893 -12.94 0.000 -.2278209 -.1678881 LOGRMEAN | .1288205 .7395089 0.17 0.862 -1.32059 1.578231 LOGR1MEAN | .337339 1.50528 0.22 0.823 -2.612956 3.287634 LOGR2MEAN | -1.058057 1.695896 -0.62 0.533 -4.381953 2.265838 LOGR3MEAN | .4736023 1.473157 0.32 0.748 -2.413731 3.360936 LOGR4MEAN | .8514155 1.125538 0.76 0.449 -1.354599 3.05743 LOGR5MEAN | -.1953429 .5483733 -0.36 0.722 -1.270135 .879449 _cons | 1.038369 .1737466 5.98 0.000 .6978321 1.378906 -------------+---------------------------------------------------------------- /lnalpha | -.2324255 .0796859 -.388607 -.076244 -------------+---------------------------------------------------------------- alpha | .7926088 .0631597 .6780007 .9265901 ------------------------------------------------------------------------------ Likelihood-ratio test of alpha=0: chibar2(01) = 2.5e+04 Prob>=chibar2 = 0.000 . estimates store PCCREdef . * Following standard errors are preferred . xtpoisson PAT $XLIST $MEANS, re vce(boot, reps($BREPS) seed(10101)) (running xtpoisson on estimation sample) Bootstrap replications (400) ----+--- 1 ---+--- 2 ---+--- 3 ---+--- 4 ---+--- 5 .................................................. 50 .................................................. 100 .................................................. 150 .................................................. 200 .................................................. 250 .................................................. 300 .................................................. 350 .................................................. 400 Random-effects Poisson regression Number of obs = 1730 Group variable: id Number of groups = 346 Random effects u_i ~ Gamma Obs per group: min = 5 avg = 5.0 max = 5 Wald chi2(18) = 1338.07 Log likelihood = -5211.898 Prob > chi2 = 0.0000 (Replications based on 346 clusters in id) ------------------------------------------------------------------------------ | Observed Bootstrap Normal-based PAT | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- LOGR | .3216624 .0893443 3.60 0.000 .1465509 .496774 LOGR1 | -.0871244 .0759254 -1.15 0.251 -.2359355 .0616866 LOGR2 | .0788715 .0646599 1.22 0.223 -.0478595 .2056025 LOGR3 | .0004448 .0823245 0.01 0.996 -.1609083 .1617978 LOGR4 | -.0047836 .0694973 -0.07 0.945 -.1409959 .1314287 LOGR5 | .0024365 .080477 0.03 0.976 -.1552955 .1601685 LOGK | .0617355 .0604269 1.02 0.307 -.0566991 .1801701 SCISECT | -.0489817 .1213592 -0.40 0.686 -.2868415 .188878 dyear2 | -.0425778 .0175546 -2.43 0.015 -.0769841 -.0081715 dyear3 | -.0399678 .0273004 -1.46 0.143 -.0934757 .0135401 dyear4 | -.1570141 .0374605 -4.19 0.000 -.2304353 -.0835928 dyear5 | -.1978545 .0399262 -4.96 0.000 -.2761084 -.1196006 LOGRMEAN | .1288205 .8918764 0.14 0.885 -1.619225 1.876866 LOGR1MEAN | .337339 1.908554 0.18 0.860 -3.403358 4.078036 LOGR2MEAN | -1.058057 2.153485 -0.49 0.623 -5.278811 3.162696 LOGR3MEAN | .4736023 1.778131 0.27 0.790 -3.01147 3.958675 LOGR4MEAN | .8514155 1.306418 0.65 0.515 -1.709117 3.411948 LOGR5MEAN | -.1953429 .5789217 -0.34 0.736 -1.330009 .9393229 _cons | 1.038369 .1958001 5.30 0.000 .6546079 1.42213 -------------+---------------------------------------------------------------- /lnalpha | -.2324255 .0872308 -.4033948 -.0614563 -------------+---------------------------------------------------------------- alpha | .7926088 .0691399 .6680483 .9403941 ------------------------------------------------------------------------------ Likelihood-ratio test of alpha=0: chibar2(01) = 2.5e+04 Prob>=chibar2 = 0.000 . estimates store PCCRErob . display "Table 9.3: fourth column Sum ln R" Table 9.3: fourth column Sum ln R . lincom LOGR + LOGR1 + LOGR2 + LOGR3 + LOGR4 + LOGR5 ( 1) [PAT]LOGR + [PAT]LOGR1 + [PAT]LOGR2 + [PAT]LOGR3 + [PAT]LOGR4 + [PAT]LOGR5 = 0 ------------------------------------------------------------------------------ PAT | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- (1) | .3115071 .1451475 2.15 0.032 .0270233 .595991 ------------------------------------------------------------------------------ . . *** TABLE 9.3: RANDOM EFFECTS - Poisson-gamma, Poisson-normal, NB, CCRE . * For Poisson RE - normal default se's given here to speed up program . * See the panel kackknife above for correct se's . estimates table PRErob PRENdef NBRErob PCCRErob, b(%11.4f) se(%11.3f) stats(N ll) ---------------------------------------------------------------------- Variable | PRErob PRENdef NBRErob PCCRErob -------------+-------------------------------------------------------- PAT | LOGR | 0.4035 0.4153 0.3503 0.3217 | 0.081 0.044 0.072 0.089 LOGR1 | -0.0462 -0.0403 -0.0030 -0.0871 | 0.077 0.048 0.072 0.076 LOGR2 | 0.1079 0.1121 0.1050 0.0789 | 0.064 0.045 0.058 0.065 LOGR3 | 0.0298 0.0348 0.0164 0.0004 | 0.084 0.041 0.077 0.082 LOGR4 | 0.0107 0.0127 0.0359 -0.0048 | 0.067 0.038 0.059 0.069 LOGR5 | 0.0406 0.0471 0.0718 0.0024 | 0.076 0.032 0.061 0.080 LOGK | 0.2917 0.2917 0.1619 0.0617 | 0.077 0.042 0.054 0.060 SCISECT | 0.2570 0.4435 0.1176 -0.0490 | 0.136 0.124 0.139 0.121 dyear2 | -0.0450 -0.0453 -0.0437 -0.0426 | 0.018 0.013 0.017 0.018 dyear3 | -0.0484 -0.0497 -0.0557 -0.0400 | 0.027 0.013 0.026 0.027 dyear4 | -0.1742 -0.1767 -0.1831 -0.1570 | 0.039 0.014 0.036 0.037 dyear5 | -0.2259 -0.2301 -0.2300 -0.1979 | 0.039 0.015 0.036 0.040 LOGRMEAN | 0.1288 | 0.892 LOGR1MEAN | 0.3373 | 1.909 LOGR2MEAN | -1.0581 | 2.153 LOGR3MEAN | 0.4736 | 1.778 LOGR4MEAN | 0.8514 | 1.306 LOGR5MEAN | -0.1953 | 0.579 _cons | 0.4108 -0.1513 0.8996 1.0384 | 0.226 0.169 0.215 0.196 -------------+-------------------------------------------------------- lnalpha | _cons | -0.1567 -0.2324 | 0.099 0.087 -------------+-------------------------------------------------------- lnsig2u | _cons | -0.0053 | 0.095 -------------+-------------------------------------------------------- ln_r | _cons | 0.9878 | 0.162 -------------+-------------------------------------------------------- ln_s | _cons | 0.7010 | 0.130 -------------+-------------------------------------------------------- Statistics | N | 1730 1730 1730 1730 ll | -5234.9265 -5245.0127 -4948.4944 -5211.8980 ---------------------------------------------------------------------- legend: b/se . . ********* DYNAMIC MODELS USING EXPONENTIAL FEEDBACK MODEL . . * Add PAT lagged once as regressor . * And now have just two lags of LOGR as regressors . global XLISTD PAT1 LOGR LOGR1 LOGR2 LOGK SCISECT dyear2 dyear3 dyear4 dyear5 . . * Pooled Poisson . poisson PAT $XLISTD, vce(cluster id) Iteration 0: log pseudolikelihood = -280117.7 Iteration 1: log pseudolikelihood = -137135.22 Iteration 2: log pseudolikelihood = -52736.981 Iteration 3: log pseudolikelihood = -18503.228 Iteration 4: log pseudolikelihood = -14743.431 Iteration 5: log pseudolikelihood = -14717.192 Iteration 6: log pseudolikelihood = -14717.183 Iteration 7: log pseudolikelihood = -14717.183 Poisson regression Number of obs = 1730 Wald chi2(10) = 671.99 Prob > chi2 = 0.0000 Log pseudolikelihood = -14717.183 Pseudo R2 = 0.8045 (Std. Err. adjusted for 346 clusters in id) ------------------------------------------------------------------------------ | Robust PAT | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- PAT1 | .0033814 .0006154 5.49 0.000 .0021753 .0045875 LOGR | .3590677 .1435441 2.50 0.012 .0777264 .640409 LOGR1 | -.1592754 .1004438 -1.59 0.113 -.3561416 .0375908 LOGR2 | .133313 .138718 0.96 0.337 -.1385693 .4051953 LOGK | .1831944 .0459439 3.99 0.000 .093146 .2732428 SCISECT | .288793 .1354546 2.13 0.033 .0233069 .5542792 dyear2 | -.0424498 .0380379 -1.12 0.264 -.1170026 .0321031 dyear3 | -.007469 .0367145 -0.20 0.839 -.079428 .06449 dyear4 | -.1213673 .0466182 -2.60 0.009 -.2127373 -.0299973 dyear5 | -.1104694 .0429585 -2.57 0.010 -.1946665 -.0262724 _cons | 1.329976 .2099348 6.34 0.000 .9185113 1.74144 ------------------------------------------------------------------------------ . estimates store DPCS . display "Table 9.5: first column Sum ln R" Table 9.5: first column Sum ln R . lincom LOGR + LOGR1 + LOGR2 ( 1) [PAT]LOGR + [PAT]LOGR1 + [PAT]LOGR2 = 0 ------------------------------------------------------------------------------ PAT | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- (1) | .3331053 .0617574 5.39 0.000 .2120631 .4541475 ------------------------------------------------------------------------------ . . * Population Averaged Poisson with exchangeable errors . * Following standard errors are preferred . xtpoisson PAT $XLISTD, pa vce(robust) Iteration 1: tolerance = .07257923 Iteration 2: tolerance = .01848274 Iteration 3: tolerance = .00150935 Iteration 4: tolerance = .00009856 Iteration 5: tolerance = 4.114e-06 Iteration 6: tolerance = 2.943e-07 GEE population-averaged model Number of obs = 1730 Group variable: id Number of groups = 346 Link: log Obs per group: min = 5 Family: Poisson avg = 5.0 Correlation: exchangeable max = 5 Wald chi2(10) = 507.36 Scale parameter: 1 Prob > chi2 = 0.0000 (Std. Err. adjusted for clustering on id) ------------------------------------------------------------------------------ | Robust PAT | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- PAT1 | .0018599 .0003348 5.55 0.000 .0012036 .0025161 LOGR | .3788088 .0661193 5.73 0.000 .2492173 .5084003 LOGR1 | -.0801661 .0742112 -1.08 0.280 -.2256175 .0652852 LOGR2 | .0778156 .0586969 1.33 0.185 -.0372282 .1928595 LOGK | .2228567 .0448267 4.97 0.000 .1349979 .3107154 SCISECT | .3706687 .155546 2.38 0.017 .0658041 .6755333 dyear2 | -.0439155 .0237863 -1.85 0.065 -.0905358 .0027047 dyear3 | -.0323041 .0256807 -1.26 0.208 -.0826372 .0180291 dyear4 | -.1543135 .0384815 -4.01 0.000 -.2297359 -.078891 dyear5 | -.1694393 .0349305 -4.85 0.000 -.2379018 -.1009767 _cons | 1.117729 .2222871 5.03 0.000 .6820544 1.553404 ------------------------------------------------------------------------------ . display "Table 9.5: second column Sum ln R" Table 9.5: second column Sum ln R . lincom LOGR + LOGR1 + LOGR2 ( 1) LOGR + LOGR1 + LOGR2 = 0 ------------------------------------------------------------------------------ PAT | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- (1) | .3764583 .0531352 7.08 0.000 .2723151 .4806015 ------------------------------------------------------------------------------ . . * Poisson Random Effects - gamma . estimates store DPPA . xtpoisson PAT $XLISTD, re vce(boot, reps($BREPS) seed(10101)) (running xtpoisson on estimation sample) Bootstrap replications (400) ----+--- 1 ---+--- 2 ---+--- 3 ---+--- 4 ---+--- 5 .................................................. 50 .................................................. 100 .................................................. 150 .................................................. 200 .................................................. 250 .................................................. 300 .................................................. 350 .................................................. 400 Random-effects Poisson regression Number of obs = 1730 Group variable: id Number of groups = 346 Random effects u_i ~ Gamma Obs per group: min = 5 avg = 5.0 max = 5 Wald chi2(10) = 1330.22 Log likelihood = -5188.0204 Prob > chi2 = 0.0000 (Replications based on 346 clusters in id) ------------------------------------------------------------------------------ | Observed Bootstrap Normal-based PAT | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- PAT1 | .0012825 .0005778 2.22 0.026 .0001501 .002415 LOGR | .44618 .0797825 5.59 0.000 .2898091 .6025508 LOGR1 | -.0595111 .0904428 -0.66 0.511 -.2367757 .1177535 LOGR2 | .1032655 .0660265 1.56 0.118 -.026144 .232675 LOGK | .3001179 .0553427 5.42 0.000 .1916483 .4085875 SCISECT | .2804724 .116429 2.41 0.016 .0522757 .5086691 dyear2 | -.0463529 .0221779 -2.09 0.037 -.0898207 -.0028851 dyear3 | -.0416408 .0271882 -1.53 0.126 -.0949288 .0116472 dyear4 | -.1680358 .037999 -4.42 0.000 -.2425125 -.0935591 dyear5 | -.1963685 .0360068 -5.45 0.000 -.2669405 -.1257964 _cons | .3724811 .1780802 2.09 0.036 .0234502 .7215119 -------------+---------------------------------------------------------------- /lnalpha | -.180793 .0945499 -.3661075 .0045215 -------------+---------------------------------------------------------------- alpha | .8346081 .0789122 .6934283 1.004532 ------------------------------------------------------------------------------ Likelihood-ratio test of alpha=0: chibar2(01) = 1.9e+04 Prob>=chibar2 = 0.000 . estimates store DPRE . display "Table 9.5: third column Sum ln R" Table 9.5: third column Sum ln R . lincom LOGR + LOGR1 + LOGR2 ( 1) [PAT]LOGR + [PAT]LOGR1 + [PAT]LOGR2 = 0 ------------------------------------------------------------------------------ PAT | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- (1) | .4899344 .0659783 7.43 0.000 .3606192 .6192495 ------------------------------------------------------------------------------ . . generate PAT_YEAR0 = PAT1 if YEAR==1 // = PAT1 in YEAR 1 and missing in YEARS 1-5 (1384 missing values generated) . bysort id: egen PAT_INITIAL = mean(PAT_YEAR0) // Replaces missings with PAT1 in YEAR1 . . global XLISTD2 PAT1 LOGR LOGR1 LOGR2 LOGK SCISECT PAT_INITIAL LOGRMEAN LOGR1MEAN LOGR2MEAN dyear2 dyear3 dyear4 dyear5 . . * Correlated random effects versions of the same . * Cross-section Poisson . poisson PAT $XLISTD2, vce(cluster id) Iteration 0: log pseudolikelihood = -279821.89 Iteration 1: log pseudolikelihood = -169670.44 (backed up) Iteration 2: log pseudolikelihood = -94501.083 Iteration 3: log pseudolikelihood = -18376.372 Iteration 4: log pseudolikelihood = -14744.374 Iteration 5: log pseudolikelihood = -14580.185 Iteration 6: log pseudolikelihood = -14579.679 Iteration 7: log pseudolikelihood = -14579.679 Poisson regression Number of obs = 1730 Wald chi2(14) = 764.86 Prob > chi2 = 0.0000 Log pseudolikelihood = -14579.679 Pseudo R2 = 0.8063 (Std. Err. adjusted for 346 clusters in id) ------------------------------------------------------------------------------ | Robust PAT | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- PAT1 | .004659 .0013789 3.38 0.001 .0019563 .0073616 LOGR | .3680191 .1265515 2.91 0.004 .1199827 .6160554 LOGR1 | -.1761849 .1331148 -1.32 0.186 -.4370851 .0847152 LOGR2 | -.0392388 .0780056 -0.50 0.615 -.192127 .1136494 LOGK | .1720819 .0453834 3.79 0.000 .0831321 .2610317 SCISECT | .2719629 .1217646 2.23 0.026 .0333086 .5106172 PAT_INITIAL | -.0014259 .0011071 -1.29 0.198 -.0035959 .000744 LOGRMEAN | .5861149 1.094472 0.54 0.592 -1.559011 2.731241 LOGR1MEAN | -1.556635 1.834338 -0.85 0.396 -5.151871 2.038601 LOGR2MEAN | 1.178711 .8811638 1.34 0.181 -.5483385 2.90576 dyear2 | -.0402669 .0413444 -0.97 0.330 -.1213004 .0407666 dyear3 | -.0022761 .038042 -0.06 0.952 -.0768371 .0722849 dyear4 | -.1118548 .0457641 -2.44 0.015 -.2015509 -.0221588 dyear5 | -.0772369 .0404438 -1.91 0.056 -.1565052 .0020315 _cons | 1.334279 .2061876 6.47 0.000 .9301587 1.738399 ------------------------------------------------------------------------------ . estimates store DPCS2 . * Population averaged Poisson with exchangeable errrors . xtpoisson PAT $XLISTD2, pa vce(robust) Iteration 1: tolerance = .63412383 Iteration 2: tolerance = .13295781 Iteration 3: tolerance = .00696397 Iteration 4: tolerance = .00007082 Iteration 5: tolerance = .00006766 Iteration 6: tolerance = 5.306e-06 Iteration 7: tolerance = 7.562e-07 GEE population-averaged model Number of obs = 1730 Group variable: id Number of groups = 346 Link: log Obs per group: min = 5 Family: Poisson avg = 5.0 Correlation: exchangeable max = 5 Wald chi2(14) = 544.01 Scale parameter: 1 Prob > chi2 = 0.0000 (Std. Err. adjusted for clustering on id) ------------------------------------------------------------------------------ | Robust PAT | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- PAT1 | .0013081 .0003492 3.75 0.000 .0006237 .0019925 LOGR | .3124704 .0680773 4.59 0.000 .1790413 .4458995 LOGR1 | -.1008425 .0717563 -1.41 0.160 -.2414823 .0397973 LOGR2 | .0454956 .0597698 0.76 0.447 -.071651 .1626421 LOGK | .1770891 .0540251 3.28 0.001 .0712018 .2829764 SCISECT | .3222529 .1679943 1.92 0.055 -.0070098 .6515157 PAT_INITIAL | .0016659 .0006676 2.50 0.013 .0003575 .0029743 LOGRMEAN | -.2157091 1.082218 -0.20 0.842 -2.336817 1.905399 LOGR1MEAN | .2100467 1.899797 0.11 0.912 -3.513487 3.93358 LOGR2MEAN | .0995838 .9866408 0.10 0.920 -1.834197 2.033364 dyear2 | -.0417337 .020654 -2.02 0.043 -.0822148 -.0012526 dyear3 | -.0308182 .0244092 -1.26 0.207 -.0786593 .0170229 dyear4 | -.1443294 .0349789 -4.13 0.000 -.2128867 -.0757721 dyear5 | -.15899 .0333839 -4.76 0.000 -.2244212 -.0935588 _cons | 1.360092 .2377173 5.72 0.000 .8941745 1.826009 ------------------------------------------------------------------------------ . estimates store DPPA2 . . * Poisson Random Effects - gamma . xtpoisson PAT $XLISTD2, re vce(boot, reps($BREPS) seed(10101)) (running xtpoisson on estimation sample) Bootstrap replications (400) ----+--- 1 ---+--- 2 ---+--- 3 ---+--- 4 ---+--- 5 .................................................. 50 .................................................. 100 .................................................. 150 .................................................. 200 .................................................. 250 .................................................. 300 .................................................. 350 .................................................. 400 Random-effects Poisson regression Number of obs = 1730 Group variable: id Number of groups = 346 Random effects u_i ~ Gamma Obs per group: min = 5 avg = 5.0 max = 5 Wald chi2(14) = 1149.44 Log likelihood = -5156.4535 Prob > chi2 = 0.0000 (Replications based on 346 clusters in id) ------------------------------------------------------------------------------ | Observed Bootstrap Normal-based PAT | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- PAT1 | .0012248 .000593 2.07 0.039 .0000626 .0023871 LOGR | .3404846 .0871984 3.90 0.000 .1695789 .5113903 LOGR1 | -.1040997 .0898571 -1.16 0.247 -.2802164 .072017 LOGR2 | .0474496 .0678818 0.70 0.485 -.0855963 .1804955 LOGK | .0447689 .0542298 0.83 0.409 -.0615195 .1510574 SCISECT | -.0397818 .1126958 -0.35 0.724 -.2606615 .181098 PAT_INITIAL | .0044761 .0015915 2.81 0.005 .0013568 .0075955 LOGRMEAN | -.1025678 .8831176 -0.12 0.908 -1.833447 1.628311 LOGR1MEAN | -.2128072 1.755537 -0.12 0.904 -3.653596 3.227981 LOGR2MEAN | .7320889 .9926132 0.74 0.461 -1.213397 2.677575 dyear2 | -.0447412 .0215512 -2.08 0.038 -.0869809 -.0025016 dyear3 | -.0372623 .0272449 -1.37 0.171 -.0906612 .0161367 dyear4 | -.1535682 .0372513 -4.12 0.000 -.2265794 -.080557 dyear5 | -.1702754 .0369497 -4.61 0.000 -.2426955 -.0978553 _cons | 1.03913 .173904 5.98 0.000 .6982845 1.379976 -------------+---------------------------------------------------------------- /lnalpha | -.3130783 .090987 -.4914094 -.1347471 -------------+---------------------------------------------------------------- alpha | .7311927 .066529 .6117636 .8739369 ------------------------------------------------------------------------------ Likelihood-ratio test of alpha=0: chibar2(01) = 1.9e+04 Prob>=chibar2 = 0.000 . estimates store DPCCRE . display "Table 9.6: first column Sum ln R" Table 9.6: first column Sum ln R . lincom LOGR + LOGR1 + LOGR2 ( 1) [PAT]LOGR + [PAT]LOGR1 + [PAT]LOGR2 = 0 ------------------------------------------------------------------------------ PAT | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- (1) | .2838345 .0960988 2.95 0.003 .0954843 .4721848 ------------------------------------------------------------------------------ . . estimates table DPCS DPCS2 DPPA DPPA2 DPRE DPCCRE, b(%7.4f) se(%7.3f) stats(N ll) stfmt(%9.1f) modelwidth(9) equations(1) -------------------------------------------------------------------------------------- Variable | DPCS DPCS2 DPPA DPPA2 DPRE DPCCRE -------------+------------------------------------------------------------------------ #1 | PAT1 | 0.0034 0.0047 0.0019 0.0013 0.0013 0.0012 | 0.001 0.001 0.000 0.000 0.001 0.001 LOGR | 0.3591 0.3680 0.3788 0.3125 0.4462 0.3405 | 0.144 0.127 0.066 0.068 0.080 0.087 LOGR1 | -0.1593 -0.1762 -0.0802 -0.1008 -0.0595 -0.1041 | 0.100 0.133 0.074 0.072 0.090 0.090 LOGR2 | 0.1333 -0.0392 0.0778 0.0455 0.1033 0.0474 | 0.139 0.078 0.059 0.060 0.066 0.068 LOGK | 0.1832 0.1721 0.2229 0.1771 0.3001 0.0448 | 0.046 0.045 0.045 0.054 0.055 0.054 SCISECT | 0.2888 0.2720 0.3707 0.3223 0.2805 -0.0398 | 0.135 0.122 0.156 0.168 0.116 0.113 dyear2 | -0.0424 -0.0403 -0.0439 -0.0417 -0.0464 -0.0447 | 0.038 0.041 0.024 0.021 0.022 0.022 dyear3 | -0.0075 -0.0023 -0.0323 -0.0308 -0.0416 -0.0373 | 0.037 0.038 0.026 0.024 0.027 0.027 dyear4 | -0.1214 -0.1119 -0.1543 -0.1443 -0.1680 -0.1536 | 0.047 0.046 0.038 0.035 0.038 0.037 dyear5 | -0.1105 -0.0772 -0.1694 -0.1590 -0.1964 -0.1703 | 0.043 0.040 0.035 0.033 0.036 0.037 PAT_INITIAL | -0.0014 0.0017 0.0045 | 0.001 0.001 0.002 LOGRMEAN | 0.5861 -0.2157 -0.1026 | 1.094 1.082 0.883 LOGR1MEAN | -1.5566 0.2100 -0.2128 | 1.834 1.900 1.756 LOGR2MEAN | 1.1787 0.0996 0.7321 | 0.881 0.987 0.993 _cons | 1.3300 1.3343 1.1177 1.3601 0.3725 1.0391 | 0.210 0.206 0.222 0.238 0.178 0.174 -------------+------------------------------------------------------------------------ lnalpha | _cons | -0.1808 -0.3131 | 0.095 0.091 -------------+------------------------------------------------------------------------ Statistics | N | 1730 1730 1730 1730 1730 1730 ll | -14717.2 -14579.7 -5188.0 -5156.5 -------------------------------------------------------------------------------------- legend: b/se . . * Fixed effects GMM using Chamberlain transformation . * This program is the same as gmm_poipre in Stata manual [r]gmm . program gmm_poipre 1. version 11 2. syntax varlist if, at(name) myrhs(varlist) /// > mylhs(varlist) myidvar(varlist) 3. quietly { 4. tempvar mu mubar ybar 5. gen double `mu' = 0 `if' 6. local j = 1 7. foreach var of varlist `myrhs' { 8. replace `mu' = `mu' + `var'*`at'[1,`j'] `if' 9. local j = `j' + 1 10. } 11. replace `mu' = exp(`mu') 12. replace `varlist' = L.`mylhs' - L.`mu'*`mylhs'/`mu' `if' 13. } 14. end . . * Only include time-varying regressors . * Also here year 1 is dropped, so drop the year 2 dummy . * Regressors . global XLISTTV PAT1 LOGR LOGR1 LOGR2 dyear3 dyear4 dyear5 . * Instruments in just-identified case . global IVLISTTV PAT2 LOGR1 LOGR2 LOGR3 dyear3 dyear4 dyear5 . * Instruments in over-identified case . global IVLISTTV2 PAT2 PAT3 PAT4 LOGR1 LOGR2 LOGR3 dyear3 dyear4 dyear5 . . * Just-identified . gmm gmm_poipre, mylhs(PAT) myrhs($XLISTTV) myidvar(id) nequations(1) /// > parameters($XLISTTV) instruments($IVLISTTV, noconstant) onestep vce(cluster id) warning: 346 missing values returned for equation 1 at initial values Step 1 Iteration 0: GMM criterion Q(b) = 9.9319006 Iteration 1: GMM criterion Q(b) = 5.1389865 Iteration 2: GMM criterion Q(b) = .39579292 Iteration 3: GMM criterion Q(b) = .00177077 Iteration 4: GMM criterion Q(b) = 1.610e-06 Iteration 5: GMM criterion Q(b) = 1.158e-12 Iteration 6: GMM criterion Q(b) = 6.158e-24 GMM estimation Number of parameters = 7 Number of moments = 7 Initial weight matrix: Unadjusted Number of obs = 1384 (Std. Err. adjusted for 346 clusters in id) ------------------------------------------------------------------------------ | Robust | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- /PAT1 | .001566 .0017031 0.92 0.358 -.0017721 .0049041 /LOGR | 1.659566 .5813948 2.85 0.004 .5200527 2.799079 /LOGR1 | -.2220653 .1397524 -1.59 0.112 -.495975 .0518444 /LOGR2 | .1711975 .0856702 2.00 0.046 .003287 .3391081 /dyear3 | -.0245858 .0281413 -0.87 0.382 -.0797416 .0305701 /dyear4 | -.2123264 .0521338 -4.07 0.000 -.3145068 -.1101459 /dyear5 | -.2947 .0734972 -4.01 0.000 -.4387518 -.1506482 ------------------------------------------------------------------------------ Instruments for equation 1: PAT2 LOGR1 LOGR2 LOGR3 dyear3 dyear4 dyear5 . estimates store DPGMM . . * Overidentified . gmm gmm_poipre, mylhs(PAT) myrhs($XLISTTV) myidvar(id) nequations(1) /// > parameters($XLISTTV) instruments($IVLISTTV2, noconstant) twostep vce(cluster id) warning: 346 missing values returned for equation 1 at initial values Step 1 Iteration 0: GMM criterion Q(b) = 14.292883 Iteration 1: GMM criterion Q(b) = 5.4360977 Iteration 2: GMM criterion Q(b) = 5.3031323 Iteration 3: GMM criterion Q(b) = 5.2945936 Iteration 4: GMM criterion Q(b) = 5.2924248 Iteration 5: GMM criterion Q(b) = 5.2919377 Iteration 6: GMM criterion Q(b) = 5.291809 Iteration 7: GMM criterion Q(b) = 5.2917771 Iteration 8: GMM criterion Q(b) = 5.2917688 Iteration 9: GMM criterion Q(b) = 5.2917667 Iteration 10: GMM criterion Q(b) = 5.2917662 Step 2 Iteration 0: GMM criterion Q(b) = .00864395 Iteration 1: GMM criterion Q(b) = .00306132 Iteration 2: GMM criterion Q(b) = .00298754 Iteration 3: GMM criterion Q(b) = .00297993 Iteration 4: GMM criterion Q(b) = .00297973 Iteration 5: GMM criterion Q(b) = .00297973 GMM estimation Number of parameters = 7 Number of moments = 9 Initial weight matrix: Unadjusted Number of obs = 1384 GMM weight matrix: Cluster (id) (Std. Err. adjusted for 346 clusters in id) ------------------------------------------------------------------------------ | Robust | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- /PAT1 | -.0001445 .0014953 -0.10 0.923 -.0030753 .0027863 /LOGR | .3003946 .7999755 0.38 0.707 -1.267529 1.868318 /LOGR1 | -.0681041 .1096623 -0.62 0.535 -.2830383 .1468301 /LOGR2 | .1324772 .0775931 1.71 0.088 -.0196025 .2845569 /dyear3 | .0093248 .0412341 0.23 0.821 -.0714925 .090142 /dyear4 | -.0954911 .0837625 -1.14 0.254 -.2596627 .0686804 /dyear5 | -.1432391 .1383765 -1.04 0.301 -.414452 .1279738 ------------------------------------------------------------------------------ Instruments for equation 1: PAT2 PAT3 PAT4 LOGR1 LOGR2 LOGR3 dyear3 dyear4 dyear5 . estimates store DPGMMOID . display "Table 9.6: second column Sum ln R" Table 9.6: second column Sum ln R . lincom _b[LOGR:_cons]+_b[LOGR1:_cons]+_b[LOGR2:_cons] ( 1) [LOGR]_cons + [LOGR1]_cons + [LOGR2]_cons = 0 ------------------------------------------------------------------------------ | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- (1) | .3647677 .7956494 0.46 0.647 -1.194676 1.924212 ------------------------------------------------------------------------------ . estat overid Test of overidentifying restriction: Hansen's J chi2(2) = 4.12395 (p = 0.1272) . predict residGMM . correlate residGMM L.residGMM L2.residGMM (obs=692) | L. L2. | residGMM residGMM residGMM -------------+--------------------------- residGMM | --. | 1.0000 L1. | -0.1736 1.0000 L2. | 0.0725 0.1071 1.0000 . . *** CHECK: THIS DOES POISSON FE USING GMM COMMAND . . * This program is the same as gmm_poi in Stata manual [r]gmm . program gmm_poi2 1. version 11 2. syntax varlist if, at(name) myrhs(varlist) /// > mylhs(varlist) myidvar(varlist) 3. quietly { 4. tempvar mu mubar ybar 5. gen double `mu' = 0 `if' 6. local j = 1 7. foreach var of varlist `myrhs' { 8. replace `mu' = `mu' + `var'*`at'[1,`j'] `if' 9. local j = `j' + 1 10. } 11. replace `mu' = exp(`mu') 12. egen double `mubar' = mean(`mu') `if', by(`myidvar') 13. egen double `ybar' = mean(`mylhs') `if', by(`myidvar') 14. replace `varlist' = `mylhs' - `mu'*`ybar'/`mubar' `if' 15. } 16. end . gmm gmm_poi2, mylhs(PAT) myrhs($XLISTTIMEVARYING) /// > myidvar(id) nequations(1) parameters($XLISTTIMEVARYING) /// > instruments($XLISTTIMEVARYING, noconstant) onestep vce(cluster id) Step 1 Iteration 0: GMM criterion Q(b) = 2.9813256 Iteration 1: GMM criterion Q(b) = .00139024 Iteration 2: GMM criterion Q(b) = 5.343e-10 Iteration 3: GMM criterion Q(b) = 1.850e-22 GMM estimation Number of parameters = 10 Number of moments = 10 Initial weight matrix: Unadjusted Number of obs = 1730 (Std. Err. adjusted for 346 clusters in id) ------------------------------------------------------------------------------ | Robust | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- /LOGR | .3222105 .0807547 3.99 0.000 .1639341 .4804868 /LOGR1 | -.0871295 .0712049 -1.22 0.221 -.2266885 .0524295 /LOGR2 | .0785816 .0620597 1.27 0.205 -.0430532 .2002164 /LOGR3 | .00106 .078183 0.01 0.989 -.1521758 .1542958 /LOGR4 | -.0046414 .063583 -0.07 0.942 -.1292617 .119979 /LOGR5 | .0026068 .0759235 0.03 0.973 -.1462004 .1514141 /dyear2 | -.0426076 .0167407 -2.55 0.011 -.0754187 -.0097965 /dyear3 | -.0400462 .0248168 -1.61 0.107 -.0886862 .0085939 /dyear4 | -.1571185 .035894 -4.38 0.000 -.2274694 -.0867676 /dyear5 | -.1980306 .0368759 -5.37 0.000 -.2703059 -.1257552 ------------------------------------------------------------------------------ Instruments for equation 1: LOGR LOGR1 LOGR2 LOGR3 LOGR4 LOGR5 dyear2 dyear3 dyear4 dyear5 . estimates store PFEGMM . . ************** CROSS-SECTION SUMMARY . . estimates table PCSdef PCSrob PCSclu NBCSdef NBCSrob NBCSclu, b(%7.4f) se(%7.3f) stats(N ll) stfmt(%9.1f) modelwidth(9) -------------------------------------------------------------------------------------- Variable | PCSdef PCSrob PCSclu NBCSdef NBCSrob NBCSclu -------------+------------------------------------------------------------------------ PAT | LOGR | 0.1345 0.1345 0.1345 0.4311 0.4311 0.4311 | 0.031 0.180 0.183 0.112 0.141 0.133 LOGR1 | -0.0529 -0.0529 -0.0529 -0.1171 -0.1171 -0.1171 | 0.043 0.242 0.106 0.156 0.186 0.141 LOGR2 | 0.0082 0.0082 0.0082 0.1065 0.1065 0.1065 | 0.040 0.232 0.093 0.150 0.168 0.121 LOGR3 | 0.0661 0.0661 0.0661 0.0764 0.0764 0.0764 | 0.037 0.221 0.114 0.137 0.155 0.103 LOGR4 | 0.0902 0.0902 0.0902 0.1938 0.1938 0.1938 | 0.033 0.198 0.093 0.125 0.128 0.088 LOGR5 | 0.2395 0.2395 0.2395 0.1194 0.1194 0.1194 | 0.022 0.132 0.123 0.085 0.090 0.086 LOGK | 0.2529 0.2529 0.2529 0.1013 0.1013 0.1013 | 0.004 0.028 0.059 0.024 0.027 0.054 SCISECT | 0.4543 0.4543 0.4543 -0.0046 -0.0046 -0.0046 | 0.009 0.077 0.167 0.056 0.059 0.119 dyear2 | -0.0435 -0.0435 -0.0435 -0.0558 -0.0558 -0.0558 | 0.013 0.096 0.018 0.077 0.076 0.035 dyear3 | -0.0524 -0.0524 -0.0524 -0.0609 -0.0609 -0.0609 | 0.013 0.097 0.030 0.077 0.080 0.043 dyear4 | -0.1702 -0.1702 -0.1702 -0.1220 -0.1220 -0.1220 | 0.014 0.094 0.046 0.077 0.085 0.047 dyear5 | -0.2019 -0.2019 -0.2019 -0.2267 -0.2267 -0.2267 | 0.014 0.089 0.046 0.077 0.085 0.049 _cons | 0.8099 0.8099 0.8099 0.9088 0.9088 0.9088 | 0.021 0.130 0.242 0.097 0.105 0.182 -------------+------------------------------------------------------------------------ lnalpha | _cons | -0.2660 -0.2660 -0.2660 | 0.044 0.048 0.089 -------------+------------------------------------------------------------------------ Statistics | N | 1730 1730 1730 1730 1730 1730 ll | -17834.1 -17834.1 -17834.1 -5773.4 -5773.4 -5773.4 -------------------------------------------------------------------------------------- legend: b/se . . ************** POPULATION AVERAGED SUMMARY . . estimates table PPAEXdef PPAEXrob PPAARrob NBPAEXdef NBPAEXrob NBPAARrob, b(%7.4f) se(%7.3f) stats(N ll) stfmt(%9.1f) modelwi > dth(9) -------------------------------------------------------------------------------------- Variable | PPAEXdef PPAEXrob PPAARrob NBPAEXdef NBPAEXrob NBPAARrob -------------+------------------------------------------------------------------------ LOGR | 0.3156 0.3156 0.2573 0.5204 0.5204 0.4628 | 0.014 0.062 0.056 0.079 0.108 0.106 LOGR1 | -0.0522 -0.0522 -0.0280 -0.0848 -0.0848 -0.0573 | 0.016 0.060 0.058 0.093 0.113 0.105 LOGR2 | 0.1048 0.1048 0.1201 0.1249 0.1249 0.1466 | 0.014 0.054 0.058 0.087 0.085 0.096 LOGR3 | 0.0197 0.0197 0.0459 0.0534 0.0534 0.0759 | 0.013 0.067 0.060 0.081 0.099 0.103 LOGR4 | 0.0230 0.0230 0.0537 0.0914 0.0914 0.1448 | 0.012 0.054 0.050 0.074 0.083 0.074 LOGR5 | 0.0489 0.0489 0.0268 0.0236 0.0236 0.0142 | 0.010 0.055 0.046 0.057 0.061 0.064 LOGK | 0.2699 0.2699 0.2623 0.1635 0.1635 0.1191 | 0.008 0.057 0.054 0.041 0.047 0.049 SCISECT | 0.4402 0.4402 0.4708 0.0687 0.0687 0.0114 | 0.019 0.175 0.170 0.105 0.114 0.115 dyear2 | -0.0456 -0.0456 -0.0453 -0.0536 -0.0536 -0.0535 | 0.005 0.017 0.017 0.045 0.034 0.034 dyear3 | -0.0462 -0.0462 -0.0447 -0.0575 -0.0575 -0.0547 | 0.005 0.026 0.026 0.045 0.040 0.042 dyear4 | -0.1686 -0.1686 -0.1633 -0.1237 -0.1237 -0.1148 | 0.005 0.041 0.042 0.045 0.046 0.048 dyear5 | -0.2136 -0.2136 -0.2053 -0.2393 -0.2393 -0.2302 | 0.005 0.041 0.041 0.045 0.049 0.052 _cons | 0.7774 0.7774 0.7498 0.7333 0.7333 0.8553 | 0.039 0.245 0.234 0.156 0.169 0.175 -------------+------------------------------------------------------------------------ N | 1730 1730 1730 1730 1730 1730 ll | -------------------------------------------------------------------------------------- legend: b/se . . * estimates table PCSdef PCSrob PCSclu PAdef PArob, b(%7.4f) se stats(N ll) star(0.05, 0.01, 0.001) equations(1) stfmt(%9.1f) > modelwidth(9) . . ************** FIXED EFFECTS SUMMARY . . estimates table PFEdef PFErob NBFEdef, b(%7.4f) se stats(N ll) equations(1) stfmt(%9.1f) modelwidth(9) -------------------------------------------------- Variable | PFEdef PFErob NBFEdef -------------+------------------------------------ LOGR | 0.3222 0.3222 0.2727 | 0.0459 0.0808 0.0708 LOGR1 | -0.0871 -0.0871 -0.0979 | 0.0487 0.0712 0.0768 LOGR2 | 0.0786 0.0786 0.0321 | 0.0448 0.0621 0.0709 LOGR3 | 0.0011 0.0011 -0.0204 | 0.0414 0.0782 0.0658 LOGR4 | -0.0046 -0.0046 0.0162 | 0.0378 0.0636 0.0629 LOGR5 | 0.0026 0.0026 -0.0097 | 0.0323 0.0759 0.0533 dyear2 | -0.0426 -0.0426 -0.0384 | 0.0131 0.0167 0.0245 dyear3 | -0.0400 -0.0400 -0.0399 | 0.0135 0.0248 0.0252 dyear4 | -0.1571 -0.1571 -0.1443 | 0.0142 0.0359 0.0265 dyear5 | -0.1980 -0.1980 -0.1958 | 0.0153 0.0369 0.0272 LOGK | 0.2071 | 0.0780 SCISECT | 0.0176 | 0.1981 _cons | 1.6614 | 0.3436 -------------+------------------------------------ N | 1620 1620 1620 ll | -3536.3 -3536.3 -3203.1 -------------------------------------------------- legend: b/se . . ************** RANDOM EFFECTS SUMMARY . . estimates table PREdef PRErob PRENdef NBREdef NBRErob, b(%7.4f) se(%7.3f) stats(N ll) stfmt(%9.1f) modelwidth(9) -------------------------------------------------------------------------- Variable | PREdef PRErob PRENdef NBREdef NBRErob -------------+------------------------------------------------------------ PAT | LOGR | 0.4035 0.4035 0.4153 0.3503 0.3503 | 0.044 0.081 0.044 0.065 0.072 LOGR1 | -0.0462 -0.0462 -0.0403 -0.0030 -0.0030 | 0.048 0.077 0.048 0.075 0.072 LOGR2 | 0.1079 0.1079 0.1121 0.1050 0.1050 | 0.045 0.064 0.045 0.069 0.058 LOGR3 | 0.0298 0.0298 0.0348 0.0164 0.0164 | 0.041 0.084 0.041 0.064 0.077 LOGR4 | 0.0107 0.0107 0.0127 0.0359 0.0359 | 0.038 0.067 0.038 0.059 0.059 LOGR5 | 0.0406 0.0406 0.0471 0.0718 0.0718 | 0.032 0.076 0.032 0.048 0.061 LOGK | 0.2917 0.2917 0.2917 0.1619 0.1619 | 0.039 0.077 0.042 0.042 0.054 SCISECT | 0.2570 0.2570 0.4435 0.1176 0.1176 | 0.112 0.136 0.124 0.107 0.139 dyear2 | -0.0450 -0.0450 -0.0453 -0.0437 -0.0437 | 0.013 0.018 0.013 0.021 0.017 dyear3 | -0.0484 -0.0484 -0.0497 -0.0557 -0.0557 | 0.013 0.027 0.013 0.022 0.026 dyear4 | -0.1742 -0.1742 -0.1767 -0.1831 -0.1831 | 0.014 0.039 0.014 0.023 0.036 dyear5 | -0.2259 -0.2259 -0.2301 -0.2300 -0.2300 | 0.015 0.039 0.015 0.023 0.036 _cons | 0.4108 0.4108 -0.1513 0.8996 0.8996 | 0.147 0.226 0.169 0.168 0.215 -------------+------------------------------------------------------------ lnalpha | _cons | -0.1567 -0.1567 | 0.081 0.099 -------------+------------------------------------------------------------ lnsig2u | _cons | -0.0053 | 0.095 -------------+------------------------------------------------------------ ln_r | _cons | 0.9878 0.9878 | 0.096 0.162 -------------+------------------------------------------------------------ ln_s | _cons | 0.7010 0.7010 | 0.108 0.130 -------------+------------------------------------------------------------ Statistics | N | 1730 1730 1730 1730 1730 ll | -5234.9 -5234.9 -5245.0 -4948.5 -4948.5 -------------------------------------------------------------------------- legend: b/se . . ************** COMBINED RESULTS SUMMARY . . * Poisson . estimates table PCSclu PPAEXrob PRErob PFErob, b(%7.4f) se(%7.3f) stats(N ll) equations(1) stfmt(%9.1f) modelwidth(9) -------------------------------------------------------------- Variable | PCSclu PPAEXrob PRErob PFErob -------------+------------------------------------------------ #1 | LOGR | 0.1345 0.3156 0.4035 0.3222 | 0.183 0.062 0.081 0.081 LOGR1 | -0.0529 -0.0522 -0.0462 -0.0871 | 0.106 0.060 0.077 0.071 LOGR2 | 0.0082 0.1048 0.1079 0.0786 | 0.093 0.054 0.064 0.062 LOGR3 | 0.0661 0.0197 0.0298 0.0011 | 0.114 0.067 0.084 0.078 LOGR4 | 0.0902 0.0230 0.0107 -0.0046 | 0.093 0.054 0.067 0.064 LOGR5 | 0.2395 0.0489 0.0406 0.0026 | 0.123 0.055 0.076 0.076 LOGK | 0.2529 0.2699 0.2917 | 0.059 0.057 0.077 SCISECT | 0.4543 0.4402 0.2570 | 0.167 0.175 0.136 dyear2 | -0.0435 -0.0456 -0.0450 -0.0426 | 0.018 0.017 0.018 0.017 dyear3 | -0.0524 -0.0462 -0.0484 -0.0400 | 0.030 0.026 0.027 0.025 dyear4 | -0.1702 -0.1686 -0.1742 -0.1571 | 0.046 0.041 0.039 0.036 dyear5 | -0.2019 -0.2136 -0.2259 -0.1980 | 0.046 0.041 0.039 0.037 _cons | 0.8099 0.7774 0.4108 | 0.242 0.245 0.226 -------------+------------------------------------------------ lnalpha | _cons | -0.1567 | 0.099 -------------+------------------------------------------------ Statistics | N | 1730 1730 1730 1620 ll | -17834.1 -5234.9 -3536.3 -------------------------------------------------------------- legend: b/se . . * Negative binomial . estimates table NBCSclu NBPAEXrob NBRErob NBFEdef, b(%7.4f) se(%7.3f) stats(N ll) equations(1) stfmt(%9.1f) modelwidth(9) -------------------------------------------------------------- Variable | NBCSclu NBPAEXrob NBRErob NBFEdef -------------+------------------------------------------------ #1 | LOGR | 0.4311 0.5204 0.3503 0.2727 | 0.133 0.108 0.072 0.071 LOGR1 | -0.1171 -0.0848 -0.0030 -0.0979 | 0.141 0.113 0.072 0.077 LOGR2 | 0.1065 0.1249 0.1050 0.0321 | 0.121 0.085 0.058 0.071 LOGR3 | 0.0764 0.0534 0.0164 -0.0204 | 0.103 0.099 0.077 0.066 LOGR4 | 0.1938 0.0914 0.0359 0.0162 | 0.088 0.083 0.059 0.063 LOGR5 | 0.1194 0.0236 0.0718 -0.0097 | 0.086 0.061 0.061 0.053 LOGK | 0.1013 0.1635 0.1619 0.2071 | 0.054 0.047 0.054 0.078 SCISECT | -0.0046 0.0687 0.1176 0.0176 | 0.119 0.114 0.139 0.198 dyear2 | -0.0558 -0.0536 -0.0437 -0.0384 | 0.035 0.034 0.017 0.024 dyear3 | -0.0609 -0.0575 -0.0557 -0.0399 | 0.043 0.040 0.026 0.025 dyear4 | -0.1220 -0.1237 -0.1831 -0.1443 | 0.047 0.046 0.036 0.026 dyear5 | -0.2267 -0.2393 -0.2300 -0.1958 | 0.049 0.049 0.036 0.027 _cons | 0.9088 0.7333 0.8996 1.6614 | 0.182 0.169 0.215 0.344 -------------+------------------------------------------------ lnalpha | _cons | -0.2660 | 0.089 -------------+------------------------------------------------ ln_r | _cons | 0.9878 | 0.162 -------------+------------------------------------------------ ln_s | _cons | 0.7010 | 0.130 -------------+------------------------------------------------ Statistics | N | 1730 1730 1730 1620 ll | -5773.4 -4948.5 -3203.1 -------------------------------------------------------------- legend: b/se . . * Poisson versus negative binomial 1 . estimates table PCSclu NBCSclu PPAEXrob NBPAEXrob, b(%7.4f) se(%7.3f) stats(N ll) equations(1) stfmt(%9.1f) modelwidth(9) -------------------------------------------------------------- Variable | PCSclu NBCSclu PPAEXrob NBPAEXrob -------------+------------------------------------------------ #1 | LOGR | 0.1345 0.4311 0.3156 0.5204 | 0.183 0.133 0.062 0.108 LOGR1 | -0.0529 -0.1171 -0.0522 -0.0848 | 0.106 0.141 0.060 0.113 LOGR2 | 0.0082 0.1065 0.1048 0.1249 | 0.093 0.121 0.054 0.085 LOGR3 | 0.0661 0.0764 0.0197 0.0534 | 0.114 0.103 0.067 0.099 LOGR4 | 0.0902 0.1938 0.0230 0.0914 | 0.093 0.088 0.054 0.083 LOGR5 | 0.2395 0.1194 0.0489 0.0236 | 0.123 0.086 0.055 0.061 LOGK | 0.2529 0.1013 0.2699 0.1635 | 0.059 0.054 0.057 0.047 SCISECT | 0.4543 -0.0046 0.4402 0.0687 | 0.167 0.119 0.175 0.114 dyear2 | -0.0435 -0.0558 -0.0456 -0.0536 | 0.018 0.035 0.017 0.034 dyear3 | -0.0524 -0.0609 -0.0462 -0.0575 | 0.030 0.043 0.026 0.040 dyear4 | -0.1702 -0.1220 -0.1686 -0.1237 | 0.046 0.047 0.041 0.046 dyear5 | -0.2019 -0.2267 -0.2136 -0.2393 | 0.046 0.049 0.041 0.049 _cons | 0.8099 0.9088 0.7774 0.7333 | 0.242 0.182 0.245 0.169 -------------+------------------------------------------------ lnalpha | _cons | -0.2660 | 0.089 -------------+------------------------------------------------ Statistics | N | 1730 1730 1730 1730 ll | -17834.1 -5773.4 -------------------------------------------------------------- legend: b/se . . * Poisson versus negative binomial 2 . estimates table PRErob NBRErob PFEdef NBFEdef, b(%7.4f) se(%7.3f) stats(N ll) equations(1) stfmt(%9.1f) modelwidth(9) -------------------------------------------------------------- Variable | PRErob NBRErob PFEdef NBFEdef -------------+------------------------------------------------ #1 | LOGR | 0.4035 0.3503 0.3222 0.2727 | 0.081 0.072 0.046 0.071 LOGR1 | -0.0462 -0.0030 -0.0871 -0.0979 | 0.077 0.072 0.049 0.077 LOGR2 | 0.1079 0.1050 0.0786 0.0321 | 0.064 0.058 0.045 0.071 LOGR3 | 0.0298 0.0164 0.0011 -0.0204 | 0.084 0.077 0.041 0.066 LOGR4 | 0.0107 0.0359 -0.0046 0.0162 | 0.067 0.059 0.038 0.063 LOGR5 | 0.0406 0.0718 0.0026 -0.0097 | 0.076 0.061 0.032 0.053 LOGK | 0.2917 0.1619 0.2071 | 0.077 0.054 0.078 SCISECT | 0.2570 0.1176 0.0176 | 0.136 0.139 0.198 dyear2 | -0.0450 -0.0437 -0.0426 -0.0384 | 0.018 0.017 0.013 0.024 dyear3 | -0.0484 -0.0557 -0.0400 -0.0399 | 0.027 0.026 0.013 0.025 dyear4 | -0.1742 -0.1831 -0.1571 -0.1443 | 0.039 0.036 0.014 0.026 dyear5 | -0.2259 -0.2300 -0.1980 -0.1958 | 0.039 0.036 0.015 0.027 _cons | 0.4108 0.8996 1.6614 | 0.226 0.215 0.344 -------------+------------------------------------------------ lnalpha | _cons | -0.1567 | 0.099 -------------+------------------------------------------------ ln_r | _cons | 0.9878 | 0.162 -------------+------------------------------------------------ ln_s | _cons | 0.7010 | 0.130 -------------+------------------------------------------------ Statistics | N | 1730 1730 1620 1620 ll | -5234.9 -4948.5 -3536.3 -3203.1 -------------------------------------------------------------- legend: b/se . . ************** TABLES in the BOOK . . *** TABLE 9.2: POOLED POISSON, POOLED GEE, POISSON FE, NB1 FE . * Note: Following gives default se's for NB1FE and not jackknife se's (given above) . estimates table PCSclu PPAEXrob PFErob NBFEdef, b(%7.4f) se(%7.3f) stats(N ll) stfmt(%9.1f) modelwidth(9) equations(1) -------------------------------------------------------------- Variable | PCSclu PPAEXrob PFErob NBFEdef -------------+------------------------------------------------ LOGR | 0.1345 0.3156 0.3222 0.2727 | 0.183 0.062 0.081 0.071 LOGR1 | -0.0529 -0.0522 -0.0871 -0.0979 | 0.106 0.060 0.071 0.077 LOGR2 | 0.0082 0.1048 0.0786 0.0321 | 0.093 0.054 0.062 0.071 LOGR3 | 0.0661 0.0197 0.0011 -0.0204 | 0.114 0.067 0.078 0.066 LOGR4 | 0.0902 0.0230 -0.0046 0.0162 | 0.093 0.054 0.064 0.063 LOGR5 | 0.2395 0.0489 0.0026 -0.0097 | 0.123 0.055 0.076 0.053 LOGK | 0.2529 0.2699 0.2071 | 0.059 0.057 0.078 SCISECT | 0.4543 0.4402 0.0176 | 0.167 0.175 0.198 dyear2 | -0.0435 -0.0456 -0.0426 -0.0384 | 0.018 0.017 0.017 0.024 dyear3 | -0.0524 -0.0462 -0.0400 -0.0399 | 0.030 0.026 0.025 0.025 dyear4 | -0.1702 -0.1686 -0.1571 -0.1443 | 0.046 0.041 0.036 0.026 dyear5 | -0.2019 -0.2136 -0.1980 -0.1958 | 0.046 0.041 0.037 0.027 _cons | 0.8099 0.7774 1.6614 | 0.242 0.245 0.344 -------------+------------------------------------------------ N | 1730 1730 1620 1620 ll | -17834.1 -3536.3 -3203.1 -------------------------------------------------------------- legend: b/se . . *** TABLE 9.3: RANDOM EFFECTS - Poisson-gamma, Poisson-normal, NB, CCRE . * For Poisson RE - normal default se's given here to speed up program (jackknife given above) . estimates table PRErob PRENdef NBRErob PCCRErob, b(%7.4f) se(%7.3f) stats(N ll) stfmt(%9.1f) modelwidth(9) equations(1) -------------------------------------------------------------- Variable | PRErob PRENdef NBRErob PCCRErob -------------+------------------------------------------------ #1 | LOGR | 0.4035 0.4153 0.3503 0.3217 | 0.081 0.044 0.072 0.089 LOGR1 | -0.0462 -0.0403 -0.0030 -0.0871 | 0.077 0.048 0.072 0.076 LOGR2 | 0.1079 0.1121 0.1050 0.0789 | 0.064 0.045 0.058 0.065 LOGR3 | 0.0298 0.0348 0.0164 0.0004 | 0.084 0.041 0.077 0.082 LOGR4 | 0.0107 0.0127 0.0359 -0.0048 | 0.067 0.038 0.059 0.069 LOGR5 | 0.0406 0.0471 0.0718 0.0024 | 0.076 0.032 0.061 0.080 LOGK | 0.2917 0.2917 0.1619 0.0617 | 0.077 0.042 0.054 0.060 SCISECT | 0.2570 0.4435 0.1176 -0.0490 | 0.136 0.124 0.139 0.121 dyear2 | -0.0450 -0.0453 -0.0437 -0.0426 | 0.018 0.013 0.017 0.018 dyear3 | -0.0484 -0.0497 -0.0557 -0.0400 | 0.027 0.013 0.026 0.027 dyear4 | -0.1742 -0.1767 -0.1831 -0.1570 | 0.039 0.014 0.036 0.037 dyear5 | -0.2259 -0.2301 -0.2300 -0.1979 | 0.039 0.015 0.036 0.040 LOGRMEAN | 0.1288 | 0.892 LOGR1MEAN | 0.3373 | 1.909 LOGR2MEAN | -1.0581 | 2.153 LOGR3MEAN | 0.4736 | 1.778 LOGR4MEAN | 0.8514 | 1.306 LOGR5MEAN | -0.1953 | 0.579 _cons | 0.4108 -0.1513 0.8996 1.0384 | 0.226 0.169 0.215 0.196 -------------+------------------------------------------------ lnalpha | _cons | -0.1567 -0.2324 | 0.099 0.087 -------------+------------------------------------------------ lnsig2u | _cons | -0.0053 | 0.095 -------------+------------------------------------------------ ln_r | _cons | 0.9878 | 0.162 -------------+------------------------------------------------ ln_s | _cons | 0.7010 | 0.130 -------------+------------------------------------------------ Statistics | N | 1730 1730 1730 1730 ll | -5234.9 -5245.0 -4948.5 -5211.9 -------------------------------------------------------------- legend: b/se . . *** TABLE 9.5 and first column TABLE 9.6: DYNAMIC MODELS RANDOM EFFECTS AND CCRE . estimates table DPCS DPPA DPRE DPCCRE, b(%7.4f) se(%7.3f) stats(N ll) stfmt(%9.1f) modelwidth(9) equations(1) -------------------------------------------------------------- Variable | DPCS DPPA DPRE DPCCRE -------------+------------------------------------------------ #1 | PAT1 | 0.0034 0.0019 0.0013 0.0012 | 0.001 0.000 0.001 0.001 LOGR | 0.3591 0.3788 0.4462 0.3405 | 0.144 0.066 0.080 0.087 LOGR1 | -0.1593 -0.0802 -0.0595 -0.1041 | 0.100 0.074 0.090 0.090 LOGR2 | 0.1333 0.0778 0.1033 0.0474 | 0.139 0.059 0.066 0.068 LOGK | 0.1832 0.2229 0.3001 0.0448 | 0.046 0.045 0.055 0.054 SCISECT | 0.2888 0.3707 0.2805 -0.0398 | 0.135 0.156 0.116 0.113 dyear2 | -0.0424 -0.0439 -0.0464 -0.0447 | 0.038 0.024 0.022 0.022 dyear3 | -0.0075 -0.0323 -0.0416 -0.0373 | 0.037 0.026 0.027 0.027 dyear4 | -0.1214 -0.1543 -0.1680 -0.1536 | 0.047 0.038 0.038 0.037 dyear5 | -0.1105 -0.1694 -0.1964 -0.1703 | 0.043 0.035 0.036 0.037 PAT_INITIAL | 0.0045 | 0.002 LOGRMEAN | -0.1026 | 0.883 LOGR1MEAN | -0.2128 | 1.756 LOGR2MEAN | 0.7321 | 0.993 _cons | 1.3300 1.1177 0.3725 1.0391 | 0.210 0.222 0.178 0.174 -------------+------------------------------------------------ lnalpha | _cons | -0.1808 -0.3131 | 0.095 0.091 -------------+------------------------------------------------ Statistics | N | 1730 1730 1730 1730 ll | -14717.2 -5188.0 -5156.5 -------------------------------------------------------------- legend: b/se . . *** TABLE 9.6: DYNAMIC MODELS FIXED EFFECTS . * Second column as first column given with TABLE 9.5 results . estimates table DPGMMOID, b(%7.4f) se(%7.3f) stats(N ll) stfmt(%9.1f) modelwidth(9) -------------------------- Variable | DPGMMOID -------------+------------ PAT1 | _cons | -0.0001 | 0.001 -------------+------------ LOGR | _cons | 0.3004 | 0.800 -------------+------------ LOGR1 | _cons | -0.0681 | 0.110 -------------+------------ LOGR2 | _cons | 0.1325 | 0.078 -------------+------------ dyear3 | _cons | 0.0093 | 0.041 -------------+------------ dyear4 | _cons | -0.0955 | 0.084 -------------+------------ dyear5 | _cons | -0.1432 | 0.138 -------------+------------ Statistics | N | 1384 ll | -------------------------- legend: b/se . . ********** CLOSE OUTPUT . . * log close . * clear . * exit . end of do-file . exit, clear