------------------------------------------------------------------------------------------------------ log: c:\Imbook\bwebpage\Section6\mma25p2matching.txt log type: text opened on: 26 May 2005, 10:26:31 . . ********** OVERVIEW OF MMA25P2MATCHING.DO ********** . . * STATA Program . * copyright C 2005 by A. Colin Cameron and Pravin K. Trivedi . * used for "Microeconometrics: Methods and Applications" . * by A. Colin Cameron and Pravin K. Trivedi (2005) . * Cambridge University Press . . * Chapter 25.8.5 pages 893-6 Tables 25.5-25.7 . * Evaluating treatment effect of training on Earnings . * using Dehejia-Wahba data (originally Lalonde data) . . * (1) For DW 2002 specification of the logit model for propensity score . * calculate treatment effect by matching methods (Tables 25.5-6) . * ( ) give distribution of propensity score (Table 25.5) . * (1A) nearest neighbor matching . * (1B) radius matching r = 0.001 . * (1C) radius matching r = 0.001 . * (1D) radius matching r = 0.001 . * (1E) stratification . * (1F) kernel matching . * (2) For DW 1999 specification of the logit model for propensity score . * calculate treatment effect by matching methods (Table 25.6) . . * The program MMA25P1TREATMENT.DO provides simpler nonmatching methods . * for the same data. . . * To run this program you need data file . * nswpsid.da1 . . * To run this program you need the Stata add-ons . * pscore.ado, atts.ado, attr.ado, attnd.ado, attnw.ado . * due to Sascha O. Becker and Andrea Ichino (2002) . * "Estimation of average treatment effects based on propensity scores", . * The Stata Journal, Vol.2, No.4, pp. 358-377. . . * This program uses version 2.02 May 13 2005 for Stata version 8 . * downloadable from http://www.iue.it/Personal/Ichino/#pscore . * We earlier used version 1.29 October 8 2002 for Stata version 7 . * downloadable from http://www.iue.it/Personal/Ichino/#pscore . * and obtained the same results . . * To speed up the program reduce breps: the number of bootstrap . * replications used to obtain bootstrap standard errors . * Bootstrap se's will differ from text as here seed is set to 10101 . . ********** STATA SETUP ********** . . set more off . version 8 . set scheme s1mono /* Used for graphs */ . . ********** DATA DESCRIPTION ********** . . * Data set nswpsid.da1 is data set nswpsid.da1 from Guido Imbens . * http://emlab.berkeley.edu/users/imbens/index.shtml . . * Data originally from DW99 . * R.H. Dehejia and S. Wahba (1999) . * "Causal Effects in Nonexperimental Studies: reevaluating the . * Evaluation of Training Programs", JASA, 1053-1062 . * or DW02 . * R.H. Dehejia and S. Wahba (2002) . * "Propensity-score Matching Methods for Nonexperimental Causal . * Studies", ReStat, 151-161 . * which in turn are from . * Lalonde, R. (1986), "Evaluating the Econometric Evaluations of . * Training Programs with Experimental Data," AER, 604-620. . . * Each observation is for an individual. . * There are 2,675 observations: 185 in treated group and 2490 in control . . * Variables are . * TREAT 1 if treated (NSW treated) and 0 if not (PSID-1 control) . * AGE in years . * EDUC in years . * BLACK 1 if black . * HISP 1 if hispanic . * MARR 1 if married . * RE74 Real annual earnings in 1974 (pre-treatment) . * RE75 Real annual earnings in 1974 (pre-treatment) . * RE78 Real annual earnings in 1974 (post-treatment) . * U74 1 if unemployed in 1974 . * U75 1 if unemployed in 1974 . . * NOTE: U74 and U75 are miscoded in these data and also in the . * summary statistics table of DW02 . * See below for correction to data . . ********** READ DATA AND TRANSFORMATIONS ********** . . ****** propensity score for nsw-psid composite sample************* . ****** output for MMA Tables 25.6 & 25.7 *********************** . . infile TREAT AGE EDUC BLACK HISP MARR RE74 RE75 RE78 U74 U75 /* > */ using nswpsid.da1 (2675 observations read) . . * The original data reversed U74 and U75 . * Should be U74=1 if R74=0 and U74=0 if R74>0 anmd similar for U75 . * This effects results with propensity score though not eariler results . . * Wrong U74 and U75 . sum U74 U75 Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- U74 | 2675 .1345794 .3413376 0 1 U75 | 2675 .1293458 .335645 0 1 . . * Correct the original data . drop U74 U75 . gen U74 = cond(RE74 == 0, 1, 0) . gen U75 = cond(RE75 == 0, 1, 0) . . * Correct U74 and U75 . sum U74 U75 Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- U74 | 2675 .1293458 .335645 0 1 U75 | 2675 .1345794 .3413376 0 1 . . * Create regressors used as additional controls in regressions below . gen AGESQ = AGE*AGE . gen EDUCSQ = EDUC*EDUC . * DW99 do not define NODEGREE but following gives Table 1 means . gen NODEGREE = 0 . replace NODEGREE = 1 if EDUC < 12 (891 real changes made) . gen RE74SQ = RE74*RE74 . gen RE75SQ = RE75*RE75 . gen U74BLACK = U74*BLACK . gen U74HISP = U74*HISP . . sum AGE EDUC NODEGREE BLACK HISP MARR U74 U75 RE74 RE75 RE78 TREAT /* > */ AGESQ EDUCSQ RE74SQ RE75SQ U74BLACK U74HISP Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- AGE | 2675 34.22579 10.49984 17 55 EDUC | 2675 11.99439 3.053556 0 17 NODEGREE | 2675 .3330841 .4714045 0 1 BLACK | 2675 .2915888 .4545789 0 1 HISP | 2675 .0343925 .1822693 0 1 -------------+-------------------------------------------------------- MARR | 2675 .8194393 .3847257 0 1 U74 | 2675 .1293458 .335645 0 1 U75 | 2675 .1345794 .3413376 0 1 RE74 | 2675 18230 13722.25 0 137149 RE75 | 2675 17850.89 13877.78 0 156653 -------------+-------------------------------------------------------- RE78 | 2675 20502.38 15632.52 0 121174 TREAT | 2675 .0691589 .2537716 0 1 AGESQ | 2675 1281.61 766.8415 289 3025 EDUCSQ | 2675 153.1862 70.62231 0 289 RE74SQ | 2675 5.21e+08 8.47e+08 0 1.88e+10 -------------+-------------------------------------------------------- RE75SQ | 2675 5.11e+08 8.91e+08 0 2.45e+10 U74BLACK | 2675 .0549533 .2279316 0 1 U74HISP | 2675 .0056075 .0746868 0 1 . . bysort TREAT: sum AGE EDUC NODEGREE BLACK HISP MARR U74 U75 RE74 RE75 RE78 TREAT /* > */ AGESQ EDUCSQ RE74SQ RE75SQ U74BLACK U74HISP ---------------------------------------------------------------------------------------------------- -> TREAT = 0 Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- AGE | 2490 34.8506 10.44076 18 55 EDUC | 2490 12.11687 3.082435 0 17 NODEGREE | 2490 .3052209 .4605934 0 1 BLACK | 2490 .2506024 .433447 0 1 HISP | 2490 .0325301 .1774389 0 1 -------------+-------------------------------------------------------- MARR | 2490 .8662651 .3404357 0 1 U74 | 2490 .0863454 .2809298 0 1 U75 | 2490 .1 .3000603 0 1 RE74 | 2490 19428.75 13406.88 0 137149 RE75 | 2490 19063.34 13596.95 0 156653 -------------+-------------------------------------------------------- RE78 | 2490 21553.92 15555.35 0 121174 TREAT | 2490 0 0 0 0 AGESQ | 2490 1323.53 769.796 324 3025 EDUCSQ | 2490 156.3161 71.43048 0 289 RE74SQ | 2490 5.57e+08 8.66e+08 0 1.88e+10 -------------+-------------------------------------------------------- RE75SQ | 2490 5.48e+08 9.12e+08 0 2.45e+10 U74BLACK | 2490 .0144578 .1193923 0 1 U74HISP | 2490 .0036145 .0600237 0 1 ---------------------------------------------------------------------------------------------------- -> TREAT = 1 Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- AGE | 185 25.81622 7.155019 17 48 EDUC | 185 10.34595 2.01065 4 16 NODEGREE | 185 .7081081 .4558666 0 1 BLACK | 185 .8432432 .3645579 0 1 HISP | 185 .0594595 .2371244 0 1 -------------+-------------------------------------------------------- MARR | 185 .1891892 .3927217 0 1 U74 | 185 .7081081 .4558666 0 1 U75 | 185 .6 .4912274 0 1 RE74 | 185 2095.574 4886.623 0 35040.1 RE75 | 185 1532.056 3219.251 0 25142.2 -------------+-------------------------------------------------------- RE78 | 185 6349.145 7867.405 0 60307.9 TREAT | 185 1 0 1 1 AGESQ | 185 717.3946 431.2517 289 2304 EDUCSQ | 185 111.0595 39.30388 16 256 RE74SQ | 185 2.81e+07 1.14e+08 0 1.23e+09 -------------+-------------------------------------------------------- RE75SQ | 185 1.27e+07 5.60e+07 0 6.32e+08 U74BLACK | 185 .6 .4912274 0 1 U74HISP | 185 .0324324 .1776263 0 1 . . *** NOTE: The benchmark estimate obtained from NSW experiment is . *** $1,794 = Average(RE_78 for NSW treated) - Average (RE_78 for NSW comtrols) . *** See MMA25P3EXTRA.DO . . ********** (1) ANALYSIS for DW02 SPECIFICATION OF THE PROPENSITY SCORE ********** . . * Following defines number of bootstrap replications . * Table 25.6 used 200 (or 100 in some places) . global breps 200 . . * From DW02 Table 3 footnote a the propensity score uses the following regressors . global XDW02 AGE AGESQ EDUC EDUCSQ MARR NODEGREE BLACK HISP RE74 RE75 RE74SQ U74 U75 U74HISP . . **** Table 25.5 p.894 summarizes propensity score . **** using just those observations with common support . . pscore TREAT $XDW02, pscore(myscore) comsup blockid(myblock) numblo(5) level(0.005) logit **************************************************** Algorithm to estimate the propensity score **************************************************** The treatment is TREAT TREAT | Freq. Percent Cum. ------------+----------------------------------- 0 | 2,490 93.08 93.08 1 | 185 6.92 100.00 ------------+----------------------------------- Total | 2,675 100.00 Estimation of the propensity score Iteration 0: log likelihood = -672.64954 Iteration 1: log likelihood = -551.87026 Iteration 2: log likelihood = -355.56578 Iteration 3: log likelihood = -234.78051 Iteration 4: log likelihood = -208.2965 Iteration 5: log likelihood = -199.26423 Iteration 6: log likelihood = -197.26114 Iteration 7: log likelihood = -197.1054 Iteration 8: log likelihood = -197.10179 Iteration 9: log likelihood = -197.10175 Logit estimates Number of obs = 2675 LR chi2(14) = 951.10 Prob > chi2 = 0.0000 Log likelihood = -197.10175 Pseudo R2 = 0.7070 ------------------------------------------------------------------------------ TREAT | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- AGE | .2628422 .120206 2.19 0.029 .0272428 .4984416 AGESQ | -.0053794 .0018341 -2.93 0.003 -.0089742 -.0017846 EDUC | .7149774 .3418173 2.09 0.036 .0450278 1.384927 EDUCSQ | -.0426178 .0179039 -2.38 0.017 -.0777088 -.0075269 MARR | -1.780857 .301802 -5.90 0.000 -2.372378 -1.189336 NODEGREE | .1891046 .4257533 0.44 0.657 -.6453564 1.023566 BLACK | 2.519383 .370358 6.80 0.000 1.793495 3.245272 HISP | 3.087327 .7340486 4.21 0.000 1.648618 4.526036 RE74 | -.0000448 .0000425 -1.05 0.292 -.000128 .0000385 RE75 | -.0002678 .0000485 -5.52 0.000 -.0003628 -.0001727 RE74SQ | 1.99e-09 7.75e-10 2.57 0.010 4.72e-10 3.51e-09 U74 | 3.100056 .5187391 5.98 0.000 2.083346 4.116766 U75 | -1.273525 .4644557 -2.74 0.006 -2.183842 -.3632088 U74HISP | -1.925803 1.07186 -1.80 0.072 -4.02661 .1750032 _cons | -7.407524 2.445692 -3.03 0.002 -12.20099 -2.614056 ------------------------------------------------------------------------------ note: 65 failures and 0 successes completely determined. Note: the common support option has been selected The region of common support is [.00036433, .98576756] Description of the estimated propensity score in region of common support Estimated propensity score ------------------------------------------------------------- Percentiles Smallest 1% .0003871 .0003643 5% .0004805 .0003669 10% .0006343 .0003702 Obs 1271 25% .0016393 .0003714 Sum of Wgt. 1271 50% .0090427 Mean .1447205 Largest Std. Dev. .2809511 75% .0897599 .9803043 90% .656286 .9830988 Variance .0789335 95% .9392306 .9855413 Skewness 2.049999 99% .9640553 .9857676 Kurtosis 5.748631 ****************************************************** Step 1: Identification of the optimal number of blocks Use option detail if you want more detailed output ****************************************************** The final number of blocks is 6 This number of blocks ensures that the mean propensity score is not different for treated and controls in each blocks ********************************************************** Step 2: Test of balancing property of the propensity score Use option detail if you want more detailed output ********************************************************** The balancing property is satisfied This table shows the inferior bound, the number of treated and the number of controls for each block Inferior | of block | TREAT of pscore | 0 1 | Total -----------+----------------------+---------- .0003643 | 960 9 | 969 .1 | 56 10 | 66 .2 | 33 14 | 47 .4 | 22 24 | 46 .6 | 7 33 | 40 .8 | 8 95 | 103 -----------+----------------------+---------- Total | 1,086 185 | 1,271 Note: the common support option has been selected ******************************************* End of the algorithm to estimate the pscore ******************************************* . . **** For completeness do same with common support option NOT selected . . drop myscore myblock . pscore TREAT $XDW02, pscore(myscore) blockid(myblock) numblo(5) level(0.005) logit **************************************************** Algorithm to estimate the propensity score **************************************************** The treatment is TREAT TREAT | Freq. Percent Cum. ------------+----------------------------------- 0 | 2,490 93.08 93.08 1 | 185 6.92 100.00 ------------+----------------------------------- Total | 2,675 100.00 Estimation of the propensity score Iteration 0: log likelihood = -672.64954 Iteration 1: log likelihood = -551.87026 Iteration 2: log likelihood = -355.56578 Iteration 3: log likelihood = -234.78051 Iteration 4: log likelihood = -208.2965 Iteration 5: log likelihood = -199.26423 Iteration 6: log likelihood = -197.26114 Iteration 7: log likelihood = -197.1054 Iteration 8: log likelihood = -197.10179 Iteration 9: log likelihood = -197.10175 Logit estimates Number of obs = 2675 LR chi2(14) = 951.10 Prob > chi2 = 0.0000 Log likelihood = -197.10175 Pseudo R2 = 0.7070 ------------------------------------------------------------------------------ TREAT | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- AGE | .2628422 .120206 2.19 0.029 .0272428 .4984416 AGESQ | -.0053794 .0018341 -2.93 0.003 -.0089742 -.0017846 EDUC | .7149774 .3418173 2.09 0.036 .0450278 1.384927 EDUCSQ | -.0426178 .0179039 -2.38 0.017 -.0777088 -.0075269 MARR | -1.780857 .301802 -5.90 0.000 -2.372378 -1.189336 NODEGREE | .1891046 .4257533 0.44 0.657 -.6453564 1.023566 BLACK | 2.519383 .370358 6.80 0.000 1.793495 3.245272 HISP | 3.087327 .7340486 4.21 0.000 1.648618 4.526036 RE74 | -.0000448 .0000425 -1.05 0.292 -.000128 .0000385 RE75 | -.0002678 .0000485 -5.52 0.000 -.0003628 -.0001727 RE74SQ | 1.99e-09 7.75e-10 2.57 0.010 4.72e-10 3.51e-09 U74 | 3.100056 .5187391 5.98 0.000 2.083346 4.116766 U75 | -1.273525 .4644557 -2.74 0.006 -2.183842 -.3632088 U74HISP | -1.925803 1.07186 -1.80 0.072 -4.02661 .1750032 _cons | -7.407524 2.445692 -3.03 0.002 -12.20099 -2.614056 ------------------------------------------------------------------------------ note: 65 failures and 0 successes completely determined. Description of the estimated propensity score Estimated propensity score ------------------------------------------------------------- Percentiles Smallest 1% 2.36e-09 1.76e-12 5% 8.39e-08 5.07e-12 10% 4.47e-07 1.14e-11 Obs 2675 25% .0000107 1.14e-11 Sum of Wgt. 2675 50% .0002558 Mean .0691589 Largest Std. Dev. .2074207 75% .0071195 .9830988 90% .129801 .9855413 Variance .0430234 95% .6394923 .9857676 Skewness 3.407447 99% .9572224 .986626 Kurtosis 13.56404 ****************************************************** Step 1: Identification of the optimal number of blocks Use option detail if you want more detailed output ****************************************************** The final number of blocks is 7 This number of blocks ensures that the mean propensity score is not different for treated and controls in each blocks ********************************************************** Step 2: Test of balancing property of the propensity score Use option detail if you want more detailed output ********************************************************** Variable BLACK is not balanced in block 1 The balancing property is not satisfied Try a different specification of the propensity score Inferior | of block | TREAT of pscore | 0 1 | Total -----------+----------------------+---------- 0 | 2,265 7 | 2,272 .05 | 98 2 | 100 .1 | 56 10 | 66 .2 | 33 14 | 47 .4 | 22 24 | 46 .6 | 7 33 | 40 .8 | 9 95 | 104 -----------+----------------------+---------- Total | 2,490 185 | 2,675 ******************************************* End of the algorithm to estimate the pscore ******************************************* . . **** All of the following use common support . . **************************************************************************** . **** Note: The results in the first half of Table 25.6 . **** erroneously added RE75SQ as a regressor. . **** This does not effect Table 25.5 (done correctly) or . **** stratification estimates (which used myscore from correct model). . **** But it does effect NN, radius and kernel estimates. . **** To enable comparison with the text we do analysis here . **** both with and without RE75SQ. . **** Even dropping RE75SQ the results continue to differ from DW02. . **** Text Corrected . **** Table 25.6 Table 25.6 DW 2002 . **** NN 2385 1286 1202 . **** Radius = 0.001 -7815 -7808 1187 . **** Radius = 0.0001 -9333 -6401 1191 . **** Radius = 0.00001 -2200 -1135 1198 . **** Stratification 1497 1497 . **** Kernel 1309 1342 . **************************************************************************** . . **** Row 1 Table 25.6: Nearest neighbor matching (random version) . set seed 10101 . attnd RE78 TREAT $XDW02 RE75SQ, comsup boot reps($breps) dots logit The program is searching the nearest neighbor of each treated unit. This operation may take a while. ATT estimation with Nearest Neighbor Matching method (random draw version) Analytical standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 185 53 2385.430 1792.028 1.331 --------------------------------------------------------- Note: the numbers of treated and controls refer to actual nearest neighbour matches Bootstrapping of standard errors command: attnd RE78 TREAT AGE AGESQ EDUC EDUCSQ MARR NODEGREE BLACK HISP RE74 RE75 RE74SQ U74 U > 75 U74HISP RE75SQ , pscore() logit comsup statistic: attnd = r(attnd) .................................................................................................... > .................................................................................................. > .. Bootstrap statistics Number of obs = 2675 Replications = 200 ------------------------------------------------------------------------------ Variable | Reps Observed Bias Std. Err. [95% Conf. Interval] -------------+---------------------------------------------------------------- attnd | 200 2385.43 -859.5093 1094.969 226.1985 4544.661 (N) | -937.0529 3515.425 (P) | 1202.547 4697.713 (BC) ------------------------------------------------------------------------------ Note: N = normal P = percentile BC = bias-corrected ATT estimation with Nearest Neighbor Matching method (random draw version) Bootstrapped standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 185 53 2385.430 1094.969 2.179 --------------------------------------------------------- Note: the numbers of treated and controls refer to actual nearest neighbour matches . set seed 10101 . attnd RE78 TREAT $XDW02, comsup boot reps($breps) dots logit The program is searching the nearest neighbor of each treated unit. This operation may take a while. ATT estimation with Nearest Neighbor Matching method (random draw version) Analytical standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 185 60 1285.782 3895.044 0.330 --------------------------------------------------------- Note: the numbers of treated and controls refer to actual nearest neighbour matches Bootstrapping of standard errors command: attnd RE78 TREAT AGE AGESQ EDUC EDUCSQ MARR NODEGREE BLACK HISP RE74 RE75 RE74SQ U74 U > 75 U74HISP , pscore() logit comsup statistic: attnd = r(attnd) .................................................................................................... > .................................................................................................. > .. Bootstrap statistics Number of obs = 2675 Replications = 200 ------------------------------------------------------------------------------ Variable | Reps Observed Bias Std. Err. [95% Conf. Interval] -------------+---------------------------------------------------------------- attnd | 200 1285.782 319.006 1275.405 -1229.261 3800.825 (N) | -1128.466 3835.567 (P) | -2181.243 3294.797 (BC) ------------------------------------------------------------------------------ Note: N = normal P = percentile BC = bias-corrected ATT estimation with Nearest Neighbor Matching method (random draw version) Bootstrapped standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 185 60 1285.782 1275.405 1.008 --------------------------------------------------------- Note: the numbers of treated and controls refer to actual nearest neighbour matches . . **** Row 2 Table 25.6: Radius matching for Radius=0.001 . set seed 10101 . attr RE78 TREAT $XDW02 RE75SQ, comsup boot reps($breps) dots logit radius(0.001) The program is searching for matches of treated units within radius. This operation may take a while. ATT estimation with the Radius Matching method Analytical standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 54 517 -7815.382 1118.181 -6.989 --------------------------------------------------------- Note: the numbers of treated and controls refer to actual matches within radius Bootstrapping of standard errors command: attr RE78 TREAT AGE AGESQ EDUC EDUCSQ MARR NODEGREE BLACK HISP RE74 RE75 RE74SQ U74 U7 > 5 U74HISP RE75SQ , pscore() logit comsup radius(.001) statistic: attr = r(attr) .................................................................................................... > .................................................................................................. > .. Bootstrap statistics Number of obs = 2675 Replications = 200 ------------------------------------------------------------------------------ Variable | Reps Observed Bias Std. Err. [95% Conf. Interval] -------------+---------------------------------------------------------------- attr | 200 -7815.381 1345.983 3794.466 -15297.9 -332.8595 (N) | -18163.96 936.3913 (P) | -21184.98 -2839.753 (BC) ------------------------------------------------------------------------------ Note: N = normal P = percentile BC = bias-corrected ATT estimation with the Radius Matching method Bootstrapped standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 54 517 -7815.381 3794.466 -2.060 --------------------------------------------------------- Note: the numbers of treated and controls refer to actual matches within radius . set seed 10101 . attr RE78 TREAT $XDW02, comsup boot reps($breps) dots logit radius(0.001) The program is searching for matches of treated units within radius. This operation may take a while. ATT estimation with the Radius Matching method Analytical standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 51 541 -7808.241 1146.418 -6.811 --------------------------------------------------------- Note: the numbers of treated and controls refer to actual matches within radius Bootstrapping of standard errors command: attr RE78 TREAT AGE AGESQ EDUC EDUCSQ MARR NODEGREE BLACK HISP RE74 RE75 RE74SQ U74 U7 > 5 U74HISP , pscore() logit comsup radius(.001) statistic: attr = r(attr) .................................................................................................... > .................................................................................................. > .. Bootstrap statistics Number of obs = 2675 Replications = 200 ------------------------------------------------------------------------------ Variable | Reps Observed Bias Std. Err. [95% Conf. Interval] -------------+---------------------------------------------------------------- attr | 200 -7808.242 1022.016 3770.093 -15242.7 -373.7819 (N) | -16697.45 1438.308 (P) | -18942.21 -1204.325 (BC) ------------------------------------------------------------------------------ Note: N = normal P = percentile BC = bias-corrected ATT estimation with the Radius Matching method Bootstrapped standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 51 541 -7808.242 3770.093 -2.071 --------------------------------------------------------- Note: the numbers of treated and controls refer to actual matches within radius . . **** Row 3 Table 25.6: Radius matching for Radius=0.0001 . set seed 10101 . attr RE78 TREAT $XDW02 RE75SQ, comsup boot reps($breps) dots logit radius(0.0001) The program is searching for matches of treated units within radius. This operation may take a while. ATT estimation with the Radius Matching method Analytical standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 24 92 -9333.120 2285.624 -4.083 --------------------------------------------------------- Note: the numbers of treated and controls refer to actual matches within radius Bootstrapping of standard errors command: attr RE78 TREAT AGE AGESQ EDUC EDUCSQ MARR NODEGREE BLACK HISP RE74 RE75 RE74SQ U74 U7 > 5 U74HISP RE75SQ , pscore() logit comsup radius(.0001) statistic: attr = r(attr) .................................................................................................... > .................................................................................................. > .. Bootstrap statistics Number of obs = 2675 Replications = 200 ------------------------------------------------------------------------------ Variable | Reps Observed Bias Std. Err. [95% Conf. Interval] -------------+---------------------------------------------------------------- attr | 200 -9333.12 4076.044 5211.11 -19609.2 942.9621 (N) | -19094.04 4604.865 (P) | -22414.52 -4341.134 (BC) ------------------------------------------------------------------------------ Note: N = normal P = percentile BC = bias-corrected ATT estimation with the Radius Matching method Bootstrapped standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 24 92 -9333.120 5211.110 -1.791 --------------------------------------------------------- Note: the numbers of treated and controls refer to actual matches within radius . set seed 10101 . attr RE78 TREAT $XDW02, comsup boot reps($breps) dots logit radius(0.0001) The program is searching for matches of treated units within radius. This operation may take a while. ATT estimation with the Radius Matching method Analytical standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 27 91 -6401.345 2054.218 -3.116 --------------------------------------------------------- Note: the numbers of treated and controls refer to actual matches within radius Bootstrapping of standard errors command: attr RE78 TREAT AGE AGESQ EDUC EDUCSQ MARR NODEGREE BLACK HISP RE74 RE75 RE74SQ U74 U7 > 5 U74HISP , pscore() logit comsup radius(.0001) statistic: attr = r(attr) .................................................................................................... > .................................................................................................. > .. Bootstrap statistics Number of obs = 2675 Replications = 200 ------------------------------------------------------------------------------ Variable | Reps Observed Bias Std. Err. [95% Conf. Interval] -------------+---------------------------------------------------------------- attr | 200 -6401.345 310.4673 5618.88 -17481.53 4678.842 (N) | -18778.71 4636.073 (P) | -21404.97 3740.767 (BC) ------------------------------------------------------------------------------ Note: N = normal P = percentile BC = bias-corrected ATT estimation with the Radius Matching method Bootstrapped standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 27 91 -6401.345 5618.880 -1.139 --------------------------------------------------------- Note: the numbers of treated and controls refer to actual matches within radius . . **** Row 4 Table 25.6: Radius matching for Radius=0.00001 . set seed 10101 . attr RE78 TREAT $XDW02 RE75SQ, comsup boot reps($breps) dots logit radius(0.00001) The program is searching for matches of treated units within radius. This operation may take a while. ATT estimation with the Radius Matching method Analytical standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 15 19 -2200.022 2986.211 -0.737 --------------------------------------------------------- Note: the numbers of treated and controls refer to actual matches within radius Bootstrapping of standard errors command: attr RE78 TREAT AGE AGESQ EDUC EDUCSQ MARR NODEGREE BLACK HISP RE74 RE75 RE74SQ U74 U7 > 5 U74HISP RE75SQ , pscore() logit comsup radius(.00001) statistic: attr = r(attr) .................................................................................................... > .................................................................................................. > .. Bootstrap statistics Number of obs = 2675 Replications = 200 ------------------------------------------------------------------------------ Variable | Reps Observed Bias Std. Err. [95% Conf. Interval] -------------+---------------------------------------------------------------- attr | 200 -2200.022 626.9762 7009.51 -16022.47 11622.43 (N) | -24355.12 8831.196 (P) | -31741.1 4217.228 (BC) ------------------------------------------------------------------------------ Note: N = normal P = percentile BC = bias-corrected ATT estimation with the Radius Matching method Bootstrapped standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 15 19 -2200.022 7009.510 -0.314 --------------------------------------------------------- Note: the numbers of treated and controls refer to actual matches within radius . set seed 10101 . attr RE78 TREAT $XDW02, comsup boot reps($breps) dots logit radius(0.00001) The program is searching for matches of treated units within radius. This operation may take a while. ATT estimation with the Radius Matching method Analytical standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 16 17 -1135.184 3189.367 -0.356 --------------------------------------------------------- Note: the numbers of treated and controls refer to actual matches within radius Bootstrapping of standard errors command: attr RE78 TREAT AGE AGESQ EDUC EDUCSQ MARR NODEGREE BLACK HISP RE74 RE75 RE74SQ U74 U7 > 5 U74HISP , pscore() logit comsup radius(.00001) statistic: attr = r(attr) .................................................................................................... > .................................................................................................. > .. Bootstrap statistics Number of obs = 2675 Replications = 200 ------------------------------------------------------------------------------ Variable | Reps Observed Bias Std. Err. [95% Conf. Interval] -------------+---------------------------------------------------------------- attr | 199 -1135.184 -2079.93 7030.204 -14998.87 12728.5 (N) | -23808.6 8048.6 (P) | -16939.85 9102.585 (BC) ------------------------------------------------------------------------------ Note: N = normal P = percentile BC = bias-corrected ATT estimation with the Radius Matching method Bootstrapped standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 16 17 -1135.184 7030.204 -0.161 --------------------------------------------------------- Note: the numbers of treated and controls refer to actual matches within radius . . **** Row 5 Table 25.6: Stratification Matching . set seed 10101 . atts RE78 TREAT, pscore(myscore) blockid(myblock) comsup boot reps($breps) dots ATT estimation with the Stratification method Analytical standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 185 1086 1497.484 920.688 1.626 --------------------------------------------------------- Bootstrapping of standard errors command: atts RE78 TREAT , pscore(myscore) blockid(myblock) comsup statistic: atts = r(atts) .................................................................................................... > .................................................................................................. > .. Bootstrap statistics Number of obs = 2675 Replications = 200 ------------------------------------------------------------------------------ Variable | Reps Observed Bias Std. Err. [95% Conf. Interval] -------------+---------------------------------------------------------------- atts | 200 1497.484 91.22797 913.129 -303.1669 3298.134 (N) | -16.69353 3509.36 (P) | -64.37524 3306.115 (BC) ------------------------------------------------------------------------------ Note: N = normal P = percentile BC = bias-corrected ATT estimation with the Stratification method Bootstrapped standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 185 1086 1497.484 913.129 1.640 --------------------------------------------------------- . . **** Row 6 Table 25.6: Kernel Matching . set seed 10101 . attk RE78 TREAT $XDW02 RE75SQ, comsup boot reps($breps) dots logit The program is searching for matches of each treated unit. This operation may take a while. ATT estimation with the Kernel Matching method --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 185 1058 1309.217 . . --------------------------------------------------------- Note: Analytical standard errors cannot be computed. Use the bootstrap option to get bootstrapped standard errors. Bootstrapping of standard errors command: attk RE78 TREAT AGE AGESQ EDUC EDUCSQ MARR NODEGREE BLACK HISP RE74 RE75 RE74SQ U74 U7 > 5 U74HISP RE75SQ , pscore() logit comsup bwidth(.06) statistic: attk = r(attk) .................................................................................................... > .................................................................................................. > .. Bootstrap statistics Number of obs = 2675 Replications = 200 ------------------------------------------------------------------------------ Variable | Reps Observed Bias Std. Err. [95% Conf. Interval] -------------+---------------------------------------------------------------- attk | 200 1309.217 45.93746 958.1801 -580.2722 3198.707 (N) | -412.7856 3416.999 (P) | -374.4567 3450.043 (BC) ------------------------------------------------------------------------------ Note: N = normal P = percentile BC = bias-corrected ATT estimation with the Kernel Matching method Bootstrapped standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 185 1058 1309.217 958.180 1.366 --------------------------------------------------------- . set seed 10101 . attk RE78 TREAT $XDW02, comsup boot reps($breps) dots logit The program is searching for matches of each treated unit. This operation may take a while. ATT estimation with the Kernel Matching method --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 185 1086 1342.016 . . --------------------------------------------------------- Note: Analytical standard errors cannot be computed. Use the bootstrap option to get bootstrapped standard errors. Bootstrapping of standard errors command: attk RE78 TREAT AGE AGESQ EDUC EDUCSQ MARR NODEGREE BLACK HISP RE74 RE75 RE74SQ U74 U7 > 5 U74HISP , pscore() logit comsup bwidth(.06) statistic: attk = r(attk) .................................................................................................... > .................................................................................................. > .. Bootstrap statistics Number of obs = 2675 Replications = 200 ------------------------------------------------------------------------------ Variable | Reps Observed Bias Std. Err. [95% Conf. Interval] -------------+---------------------------------------------------------------- attk | 200 1342.016 61.94744 933.8668 -499.5284 3183.561 (N) | -378.5027 3354.131 (P) | -405.7551 3349.118 (BC) ------------------------------------------------------------------------------ Note: N = normal P = percentile BC = bias-corrected ATT estimation with the Kernel Matching method Bootstrapped standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 185 1086 1342.016 933.867 1.437 --------------------------------------------------------- . . ********** (2) ANALYSIS for DW99 SPECIFICATION OF THE PROPENSITY SCORE ********** . . * From DW99 Table 3 footnote e the propensity score uses the following regressors . global XDW99 AGE AGESQ EDUC EDUCSQ MARR NODEGREE BLACK HISP RE74 RE75 RE74SQ RE75SQ U74BLACK . . * Note that CT Table 25.6 footnote b erroneously lists RE74*RE75 as regressor . * but this program (correctly) did not include RE74*RE75 . . **** Propensity score with just those observations with common support . . drop myscore myblock . pscore TREAT $XDW99, pscore(myscore) comsup blockid(myblock) numblo($breps) level(0.005) logit **************************************************** Algorithm to estimate the propensity score **************************************************** The treatment is TREAT TREAT | Freq. Percent Cum. ------------+----------------------------------- 0 | 2,490 93.08 93.08 1 | 185 6.92 100.00 ------------+----------------------------------- Total | 2,675 100.00 Estimation of the propensity score Iteration 0: log likelihood = -672.64954 Iteration 1: log likelihood = -499.56574 Iteration 2: log likelihood = -318.55053 Iteration 3: log likelihood = -248.28844 Iteration 4: log likelihood = -225.08984 Iteration 5: log likelihood = -219.00396 Iteration 6: log likelihood = -209.30653 Iteration 7: log likelihood = -208.38887 Iteration 8: log likelihood = -205.17689 Iteration 9: log likelihood = -204.93156 Iteration 10: log likelihood = -204.92951 Iteration 11: log likelihood = -204.9295 Logit estimates Number of obs = 2675 LR chi2(13) = 935.44 Prob > chi2 = 0.0000 Log likelihood = -204.9295 Pseudo R2 = 0.6953 ------------------------------------------------------------------------------ TREAT | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- AGE | .3305734 .1203353 2.75 0.006 .0947206 .5664262 AGESQ | -.0063429 .0018561 -3.42 0.001 -.0099808 -.0027049 EDUC | .8247711 .3534216 2.33 0.020 .1320775 1.517465 EDUCSQ | -.0483153 .0186057 -2.60 0.009 -.0847819 -.0118488 MARR | -1.884062 .2994614 -6.29 0.000 -2.470996 -1.297129 NODEGREE | .1299868 .4284278 0.30 0.762 -.7097163 .96969 BLACK | 1.132961 .352088 3.22 0.001 .4428814 1.823041 HISP | 1.962762 .5673735 3.46 0.001 .8507302 3.074793 RE74 | -.0001047 .0000355 -2.95 0.003 -.0001743 -.0000351 RE75 | -.0002172 .0000415 -5.23 0.000 -.0002986 -.0001357 RE74SQ | 2.36e-09 6.57e-10 3.59 0.000 1.07e-09 3.65e-09 RE75SQ | 1.58e-10 6.68e-10 0.24 0.813 -1.15e-09 1.47e-09 U74BLACK | 2.137042 .4273667 5.00 0.000 1.299419 2.974665 _cons | -7.552458 2.451721 -3.08 0.002 -12.35774 -2.747173 ------------------------------------------------------------------------------ note: 19 failures and 0 successes completely determined. Note: the common support option has been selected The region of common support is [.00065257, .97487544] Description of the estimated propensity score in region of common support Estimated propensity score ------------------------------------------------------------- Percentiles Smallest 1% .0006813 .0006526 5% .0008363 .0006581 10% .0011416 .0006593 Obs 1331 25% .0024351 .0006598 Sum of Wgt. 1331 50% .0111854 Mean .1388772 Largest Std. Dev. .275571 75% .0779976 .9744237 90% .6200607 .9747552 Variance .0759394 95% .9494181 .9747918 Skewness 2.17177 99% .970738 .9748754 Kurtosis 6.296349 ****************************************************** Step 1: Identification of the optimal number of blocks Use option detail if you want more detailed output ****************************************************** The final number of blocks is 195 This number of blocks ensures that the mean propensity score is not different for treated and controls in each blocks ********************************************************** Step 2: Test of balancing property of the propensity score Use option detail if you want more detailed output ********************************************************** The balancing property is satisfied This table shows the inferior bound, the number of treated and the number of controls for each block Inferior | of block | TREAT of pscore | 0 1 | Total -----------+----------------------+---------- .0006526 | 501 2 | 503 .005 | 143 3 | 146 .01 | 78 0 | 78 .015 | 42 0 | 42 .02 | 38 0 | 38 .025 | 29 1 | 30 .03 | 22 0 | 22 .035 | 23 0 | 23 .04 | 22 0 | 22 .045 | 17 1 | 18 .05 | 23 0 | 23 .055 | 13 1 | 14 .06 | 12 0 | 12 .065 | 9 0 | 9 .07 | 11 1 | 12 .075 | 9 1 | 10 .08 | 6 0 | 6 .085 | 6 0 | 6 .09 | 8 1 | 9 .095 | 6 0 | 6 .1 | 9 0 | 9 .105 | 4 0 | 4 .11 | 8 0 | 8 .115 | 3 0 | 3 .12 | 1 0 | 1 .125 | 2 3 | 5 .13 | 6 1 | 7 .135 | 1 0 | 1 .14 | 1 1 | 2 .145 | 1 0 | 1 .15 | 2 0 | 2 .155 | 4 0 | 4 .16 | 3 0 | 3 .165 | 2 0 | 2 .175 | 1 0 | 1 .18 | 0 1 | 1 .185 | 1 0 | 1 .19 | 2 0 | 2 .195 | 2 1 | 3 .2 | 1 0 | 1 .205 | 1 0 | 1 .215 | 5 0 | 5 .225 | 2 1 | 3 .23 | 2 1 | 3 .235 | 2 3 | 5 .24 | 2 0 | 2 .245 | 0 1 | 1 .25 | 0 2 | 2 .26 | 1 1 | 2 .265 | 1 0 | 1 .27 | 1 0 | 1 .28 | 1 0 | 1 .285 | 1 0 | 1 .29 | 2 1 | 3 .295 | 2 1 | 3 .3 | 2 0 | 2 .305 | 0 1 | 1 .315 | 1 0 | 1 .32 | 0 1 | 1 .325 | 2 1 | 3 .33 | 1 0 | 1 .335 | 0 1 | 1 .34 | 1 1 | 2 .345 | 1 2 | 3 .35 | 2 0 | 2 .355 | 0 1 | 1 .365 | 1 0 | 1 .37 | 2 0 | 2 .375 | 2 2 | 4 .38 | 1 2 | 3 .385 | 1 4 | 5 .4 | 0 1 | 1 .405 | 0 2 | 2 .42 | 0 1 | 1 .425 | 1 0 | 1 .45 | 2 0 | 2 .47 | 1 0 | 1 .48 | 1 1 | 2 .485 | 2 0 | 2 .495 | 1 0 | 1 .5 | 0 2 | 2 .51 | 0 2 | 2 .515 | 2 1 | 3 .525 | 0 1 | 1 .53 | 0 2 | 2 .535 | 0 1 | 1 .54 | 1 0 | 1 .555 | 0 1 | 1 .56 | 1 1 | 2 .565 | 1 0 | 1 .57 | 0 1 | 1 .575 | 1 1 | 2 .59 | 0 1 | 1 .595 | 0 1 | 1 .6 | 0 1 | 1 .605 | 0 1 | 1 .61 | 1 2 | 3 .615 | 0 1 | 1 .62 | 0 1 | 1 .625 | 0 1 | 1 .635 | 1 2 | 3 .64 | 1 1 | 2 .645 | 2 0 | 2 .665 | 0 1 | 1 .67 | 1 0 | 1 .675 | 0 3 | 3 .68 | 1 0 | 1 .69 | 1 0 | 1 .71 | 1 1 | 2 .735 | 0 1 | 1 .74 | 1 0 | 1 .745 | 2 0 | 2 .765 | 1 1 | 2 .79 | 0 4 | 4 .795 | 0 1 | 1 .8 | 0 1 | 1 .805 | 0 2 | 2 .815 | 0 3 | 3 .825 | 0 1 | 1 .84 | 0 1 | 1 .845 | 0 1 | 1 .85 | 0 1 | 1 .86 | 0 1 | 1 .865 | 0 1 | 1 .895 | 0 1 | 1 .9 | 0 2 | 2 .905 | 0 2 | 2 .915 | 0 1 | 1 .92 | 0 1 | 1 .925 | 0 7 | 7 .93 | 0 2 | 2 .935 | 0 1 | 1 .94 | 0 3 | 3 .945 | 1 6 | 7 .95 | 1 14 | 15 .955 | 0 16 | 16 .96 | 1 5 | 6 .965 | 3 12 | 15 .97 | 1 13 | 14 -----------+----------------------+---------- Total | 1,146 185 | 1,331 Note: the common support option has been selected ******************************************* End of the algorithm to estimate the pscore ******************************************* . . **** For completeness do same with common support option NOT selected . . drop myscore myblock . pscore TREAT $XDW99, pscore(myscore) blockid(myblock) numblo($breps) level(0.005) logit **************************************************** Algorithm to estimate the propensity score **************************************************** The treatment is TREAT TREAT | Freq. Percent Cum. ------------+----------------------------------- 0 | 2,490 93.08 93.08 1 | 185 6.92 100.00 ------------+----------------------------------- Total | 2,675 100.00 Estimation of the propensity score Iteration 0: log likelihood = -672.64954 Iteration 1: log likelihood = -499.56574 Iteration 2: log likelihood = -318.55053 Iteration 3: log likelihood = -248.28844 Iteration 4: log likelihood = -225.08984 Iteration 5: log likelihood = -219.00396 Iteration 6: log likelihood = -209.30653 Iteration 7: log likelihood = -208.38887 Iteration 8: log likelihood = -205.17689 Iteration 9: log likelihood = -204.93156 Iteration 10: log likelihood = -204.92951 Iteration 11: log likelihood = -204.9295 Logit estimates Number of obs = 2675 LR chi2(13) = 935.44 Prob > chi2 = 0.0000 Log likelihood = -204.9295 Pseudo R2 = 0.6953 ------------------------------------------------------------------------------ TREAT | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- AGE | .3305734 .1203353 2.75 0.006 .0947206 .5664262 AGESQ | -.0063429 .0018561 -3.42 0.001 -.0099808 -.0027049 EDUC | .8247711 .3534216 2.33 0.020 .1320775 1.517465 EDUCSQ | -.0483153 .0186057 -2.60 0.009 -.0847819 -.0118488 MARR | -1.884062 .2994614 -6.29 0.000 -2.470996 -1.297129 NODEGREE | .1299868 .4284278 0.30 0.762 -.7097163 .96969 BLACK | 1.132961 .352088 3.22 0.001 .4428814 1.823041 HISP | 1.962762 .5673735 3.46 0.001 .8507302 3.074793 RE74 | -.0001047 .0000355 -2.95 0.003 -.0001743 -.0000351 RE75 | -.0002172 .0000415 -5.23 0.000 -.0002986 -.0001357 RE74SQ | 2.36e-09 6.57e-10 3.59 0.000 1.07e-09 3.65e-09 RE75SQ | 1.58e-10 6.68e-10 0.24 0.813 -1.15e-09 1.47e-09 U74BLACK | 2.137042 .4273667 5.00 0.000 1.299419 2.974665 _cons | -7.552458 2.451721 -3.08 0.002 -12.35774 -2.747173 ------------------------------------------------------------------------------ note: 19 failures and 0 successes completely determined. Description of the estimated propensity score Estimated propensity score ------------------------------------------------------------- Percentiles Smallest 1% 2.84e-08 4.49e-11 5% 4.47e-07 4.88e-10 10% 2.07e-06 4.88e-10 Obs 2675 25% .000034 4.95e-10 Sum of Wgt. 2675 50% .0006388 Mean .0691589 Largest Std. Dev. .2063646 75% .010941 .9744237 90% .1336877 .9747552 Variance .0425863 95% .6200607 .9747918 Skewness 3.471137 99% .9651648 .9748754 Kurtosis 14.05057 ****************************************************** Step 1: Identification of the optimal number of blocks Use option detail if you want more detailed output ****************************************************** The final number of blocks is 195 This number of blocks ensures that the mean propensity score is not different for treated and controls in each blocks ********************************************************** Step 2: Test of balancing property of the propensity score Use option detail if you want more detailed output ********************************************************** Variable BLACK is not balanced in block 1 The balancing property is not satisfied Try a different specification of the propensity score Inferior | of block | TREAT of pscore | 0 1 | Total -----------+----------------------+---------- 0 | 1,845 2 | 1,847 .005 | 143 3 | 146 .01 | 78 0 | 78 .015 | 42 0 | 42 .02 | 38 0 | 38 .025 | 29 1 | 30 .03 | 22 0 | 22 .035 | 23 0 | 23 .04 | 22 0 | 22 .045 | 17 1 | 18 .05 | 23 0 | 23 .055 | 13 1 | 14 .06 | 12 0 | 12 .065 | 9 0 | 9 .07 | 11 1 | 12 .075 | 9 1 | 10 .08 | 6 0 | 6 .085 | 6 0 | 6 .09 | 8 1 | 9 .095 | 6 0 | 6 .1 | 9 0 | 9 .105 | 4 0 | 4 .11 | 8 0 | 8 .115 | 3 0 | 3 .12 | 1 0 | 1 .125 | 2 3 | 5 .13 | 6 1 | 7 .135 | 1 0 | 1 .14 | 1 1 | 2 .145 | 1 0 | 1 .15 | 2 0 | 2 .155 | 4 0 | 4 .16 | 3 0 | 3 .165 | 2 0 | 2 .175 | 1 0 | 1 .18 | 0 1 | 1 .185 | 1 0 | 1 .19 | 2 0 | 2 .195 | 2 1 | 3 .2 | 1 0 | 1 .205 | 1 0 | 1 .215 | 5 0 | 5 .225 | 2 1 | 3 .23 | 2 1 | 3 .235 | 2 3 | 5 .24 | 2 0 | 2 .245 | 0 1 | 1 .25 | 0 2 | 2 .26 | 1 1 | 2 .265 | 1 0 | 1 .27 | 1 0 | 1 .28 | 1 0 | 1 .285 | 1 0 | 1 .29 | 2 1 | 3 .295 | 2 1 | 3 .3 | 2 0 | 2 .305 | 0 1 | 1 .315 | 1 0 | 1 .32 | 0 1 | 1 .325 | 2 1 | 3 .33 | 1 0 | 1 .335 | 0 1 | 1 .34 | 1 1 | 2 .345 | 1 2 | 3 .35 | 2 0 | 2 .355 | 0 1 | 1 .365 | 1 0 | 1 .37 | 2 0 | 2 .375 | 2 2 | 4 .38 | 1 2 | 3 .385 | 1 4 | 5 .4 | 0 1 | 1 .405 | 0 2 | 2 .42 | 0 1 | 1 .425 | 1 0 | 1 .45 | 2 0 | 2 .47 | 1 0 | 1 .48 | 1 1 | 2 .485 | 2 0 | 2 .495 | 1 0 | 1 .5 | 0 2 | 2 .51 | 0 2 | 2 .515 | 2 1 | 3 .525 | 0 1 | 1 .53 | 0 2 | 2 .535 | 0 1 | 1 .54 | 1 0 | 1 .555 | 0 1 | 1 .56 | 1 1 | 2 .565 | 1 0 | 1 .57 | 0 1 | 1 .575 | 1 1 | 2 .59 | 0 1 | 1 .595 | 0 1 | 1 .6 | 0 1 | 1 .605 | 0 1 | 1 .61 | 1 2 | 3 .615 | 0 1 | 1 .62 | 0 1 | 1 .625 | 0 1 | 1 .635 | 1 2 | 3 .64 | 1 1 | 2 .645 | 2 0 | 2 .665 | 0 1 | 1 .67 | 1 0 | 1 .675 | 0 3 | 3 .68 | 1 0 | 1 .69 | 1 0 | 1 .71 | 1 1 | 2 .735 | 0 1 | 1 .74 | 1 0 | 1 .745 | 2 0 | 2 .765 | 1 1 | 2 .79 | 0 4 | 4 .795 | 0 1 | 1 .8 | 0 1 | 1 .805 | 0 2 | 2 .815 | 0 3 | 3 .825 | 0 1 | 1 .84 | 0 1 | 1 .845 | 0 1 | 1 .85 | 0 1 | 1 .86 | 0 1 | 1 .865 | 0 1 | 1 .895 | 0 1 | 1 .9 | 0 2 | 2 .905 | 0 2 | 2 .915 | 0 1 | 1 .92 | 0 1 | 1 .925 | 0 7 | 7 .93 | 0 2 | 2 .935 | 0 1 | 1 .94 | 0 3 | 3 .945 | 1 6 | 7 .95 | 1 14 | 15 .955 | 0 16 | 16 .96 | 1 5 | 6 .965 | 3 12 | 15 .97 | 1 13 | 14 -----------+----------------------+---------- Total | 2,490 185 | 2,675 ******************************************* End of the algorithm to estimate the pscore ******************************************* . . **** All of the following use common support . . **** Row 7 Table 25.6: Nearest neighbor matching (random version) . set seed 10101 . attnd RE78 TREAT $XDW99, comsup boot reps($breps) dots logit The program is searching the nearest neighbor of each treated unit. This operation may take a while. ATT estimation with Nearest Neighbor Matching method (random draw version) Analytical standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 185 57 560.287 2205.663 0.254 --------------------------------------------------------- Note: the numbers of treated and controls refer to actual nearest neighbour matches Bootstrapping of standard errors command: attnd RE78 TREAT AGE AGESQ EDUC EDUCSQ MARR NODEGREE BLACK HISP RE74 RE75 RE74SQ RE75S > Q U74BLACK , pscore() logit comsup statistic: attnd = r(attnd) .................................................................................................... > .................................................................................................. > .. Bootstrap statistics Number of obs = 2675 Replications = 200 ------------------------------------------------------------------------------ Variable | Reps Observed Bias Std. Err. [95% Conf. Interval] -------------+---------------------------------------------------------------- attnd | 200 560.2872 1104.87 1331.294 -2064.967 3185.542 (N) | -785.5272 4190.844 (P) | -2615.809 2016.239 (BC) ------------------------------------------------------------------------------ Note: N = normal P = percentile BC = bias-corrected ATT estimation with Nearest Neighbor Matching method (random draw version) Bootstrapped standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 185 57 560.287 1331.294 0.421 --------------------------------------------------------- Note: the numbers of treated and controls refer to actual nearest neighbour matches . . **** Row 8 Table 25.6: Radius matching for Radius=0.001 . set seed 10101 . attr RE78 TREAT $XDW99, comsup boot reps($breps) dots logit radius(0.001) The program is searching for matches of treated units within radius. This operation may take a while. ATT estimation with the Radius Matching method Analytical standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 57 583 -9358.228 997.561 -9.381 --------------------------------------------------------- Note: the numbers of treated and controls refer to actual matches within radius Bootstrapping of standard errors command: attr RE78 TREAT AGE AGESQ EDUC EDUCSQ MARR NODEGREE BLACK HISP RE74 RE75 RE74SQ RE75SQ > U74BLACK , pscore() logit comsup radius(.001) statistic: attr = r(attr) .................................................................................................... > .................................................................................................. > .. Bootstrap statistics Number of obs = 2675 Replications = 200 ------------------------------------------------------------------------------ Variable | Reps Observed Bias Std. Err. [95% Conf. Interval] -------------+---------------------------------------------------------------- attr | 200 -9358.228 2589.204 3079.824 -15431.51 -3284.949 (N) | -11328.39 901.8873 (P) | -13053.95 -6956.288 (BC) ------------------------------------------------------------------------------ Note: N = normal P = percentile BC = bias-corrected ATT estimation with the Radius Matching method Bootstrapped standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 57 583 -9358.228 3079.824 -3.039 --------------------------------------------------------- Note: the numbers of treated and controls refer to actual matches within radius . . **** Row 9 Table 25.6: Radius matching for Radius=0.0001 . set seed 10101 . attr RE78 TREAT $XDW99, comsup boot reps($breps) dots logit radius(0.0001) The program is searching for matches of treated units within radius. This operation may take a while. ATT estimation with the Radius Matching method Analytical standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 27 76 -7847.460 2066.697 -3.797 --------------------------------------------------------- Note: the numbers of treated and controls refer to actual matches within radius Bootstrapping of standard errors command: attr RE78 TREAT AGE AGESQ EDUC EDUCSQ MARR NODEGREE BLACK HISP RE74 RE75 RE74SQ RE75SQ > U74BLACK , pscore() logit comsup radius(.0001) statistic: attr = r(attr) .................................................................................................... > .................................................................................................. > .. Bootstrap statistics Number of obs = 2675 Replications = 200 ------------------------------------------------------------------------------ Variable | Reps Observed Bias Std. Err. [95% Conf. Interval] -------------+---------------------------------------------------------------- attr | 200 -7847.46 2920.804 4850.874 -17413.17 1718.251 (N) | -13423.91 5223.634 (P) | -15432.32 632.0693 (BC) ------------------------------------------------------------------------------ Note: N = normal P = percentile BC = bias-corrected ATT estimation with the Radius Matching method Bootstrapped standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 27 76 -7847.460 4850.874 -1.618 --------------------------------------------------------- Note: the numbers of treated and controls refer to actual matches within radius . . **** Row 10 Table 25.6: Radius matching for Radius=0.00001 . set seed 10101 . attr RE78 TREAT $XDW99, comsup boot reps($breps) dots logit radius(0.00001) The program is searching for matches of treated units within radius. This operation may take a while. ATT estimation with the Radius Matching method Analytical standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 16 13 223.468 4551.850 0.049 --------------------------------------------------------- Note: the numbers of treated and controls refer to actual matches within radius Bootstrapping of standard errors command: attr RE78 TREAT AGE AGESQ EDUC EDUCSQ MARR NODEGREE BLACK HISP RE74 RE75 RE74SQ RE75SQ > U74BLACK , pscore() logit comsup radius(.00001) statistic: attr = r(attr) .................................................................................................... > .................................................................................................. > .. Bootstrap statistics Number of obs = 2675 Replications = 200 ------------------------------------------------------------------------------ Variable | Reps Observed Bias Std. Err. [95% Conf. Interval] -------------+---------------------------------------------------------------- attr | 199 223.4685 -1272.487 5608.927 -10837.43 11284.37 (N) | -14600.21 8548.427 (P) | -10778.17 11039.05 (BC) ------------------------------------------------------------------------------ Note: N = normal P = percentile BC = bias-corrected ATT estimation with the Radius Matching method Bootstrapped standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 16 13 223.468 5608.927 0.040 --------------------------------------------------------- Note: the numbers of treated and controls refer to actual matches within radius . . **** Row 11 Table 25.6: Stratification Matching . set seed 10101 . atts RE78 TREAT, pscore(myscore) blockid(myblock) comsup boot reps($breps) dots ATT estimation with the Stratification method Analytical standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 98 1233 1322.160 . . --------------------------------------------------------- Bootstrapping of standard errors command: atts RE78 TREAT , pscore(myscore) blockid(myblock) comsup statistic: atts = r(atts) .................................................................................................... > .................................................................................................. > .. Bootstrap statistics Number of obs = 2675 Replications = 200 ------------------------------------------------------------------------------ Variable | Reps Observed Bias Std. Err. [95% Conf. Interval] -------------+---------------------------------------------------------------- atts | 200 1322.16 -51.6285 1276.237 -1194.524 3838.844 (N) | -1515.399 3960.787 (P) | -1383.034 4034.298 (BC) ------------------------------------------------------------------------------ Note: N = normal P = percentile BC = bias-corrected ATT estimation with the Stratification method Bootstrapped standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 98 1233 1322.160 1276.237 1.036 --------------------------------------------------------- . . **** Row 12 Table 25.6: Kernel Matching . * pscore TREAT $XDW99, pscore(myscore) comsup blockid(myblock) numblo($breps) level(0.005) logit . set seed 10101 . attk RE78 TREAT $XDW99, comsup boot reps($breps) dots logit The program is searching for matches of each treated unit. This operation may take a while. ATT estimation with the Kernel Matching method --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 185 1146 1518.694 . . --------------------------------------------------------- Note: Analytical standard errors cannot be computed. Use the bootstrap option to get bootstrapped standard errors. Bootstrapping of standard errors command: attk RE78 TREAT AGE AGESQ EDUC EDUCSQ MARR NODEGREE BLACK HISP RE74 RE75 RE74SQ RE75SQ > U74BLACK , pscore() logit comsup bwidth(.06) statistic: attk = r(attk) .................................................................................................... > .................................................................................................. > .. Bootstrap statistics Number of obs = 2675 Replications = 200 ------------------------------------------------------------------------------ Variable | Reps Observed Bias Std. Err. [95% Conf. Interval] -------------+---------------------------------------------------------------- attk | 200 1518.694 130.8493 808.3386 -75.31444 3112.703 (N) | 212.6286 3165.292 (P) | 96.05106 2991.407 (BC) ------------------------------------------------------------------------------ Note: N = normal P = percentile BC = bias-corrected ATT estimation with the Kernel Matching method Bootstrapped standard errors --------------------------------------------------------- n. treat. n. contr. ATT Std. Err. t --------------------------------------------------------- 185 1146 1518.694 808.339 1.879 --------------------------------------------------------- . . ********** CLOSE OUTPUT ********** . log close log: c:\Imbook\bwebpage\Section6\mma25p2matching.txt log type: text closed on: 26 May 2005, 11:15:53 ----------------------------------------------------------------------------------------------------