Doug Miller’s Stata code page
This page has Stata code used in some of my papers.
Multi-way clustering with OLS
cgmreg.ado – This is the multi-way clustering regression code used in “Robust Inference with multi-way clustering” by Cameron, Gelbach, and Miller (2006, NBER TWP 327). Some sample Stata code that uses this ado file is cgmreg_example.do .
Percentile-t cluster bootstrap example
In Cameron, Gelbach, Miller (ReStat 2008), we suggest using a “Wild cluster-bootstrap percentile-t procedure, imposing the null hypothesis,” for improved inference with few clusters. I don’t have an .ado file that will automatically run this, but I do have a .do file, which is adopted from the Gruber-Poterba Application of our paper, that can serve as an example. The file is bs_example.do, and it uses the dataset collapsed.dta. The code produces p-values for “pairs cluster bootstrap,” “cluster residual bootstrap,” and “Wild bootstrap.” (Note: The numbers for the Wild bootstrap are different than those in our paper, because in the old version of the code we used different (and less intuitive) weights for the wild bootstrap, and we did not "impose the null hypothesis". Because our base recommendation is to use the Wild Bootstrap, impose the null, and use the +1 -1 weights, I've modified the code to reflect those recommendations.)
Nonparametric Regression Discontinuity Code
An earlier version of this code was used in Ludwig and Miller (2007).
Start with the readme.txt file. The code to run is the example.do file. The actual nonparametric RD code is in the flexporter1v9.do file. An example of the graphs that can be made is in example_graph.wmf file.
Code and Public Data for “Does Head Start Improve Children’s Life Chances? Evidence From a Regression Discontinuity Design”
This .zip file (almost 7MB) when extracted contains data and Stata .do files that reproduce the tables and graphs in: Jens Ludwig and Douglas L Miller, "Does Head Start Improve Children's Life Outcomes? Evidence from a Regression Discontinuity Design," Quarterly Journal of Economics, February 2007, Vol 72, Issue 1, pp 159-208. The file "doall.do" in the folder "tables" will create all the tables. The graphs are in the folder "tables/graphs". The NELS results in the paper were based on geo-coded NELS data, and this data is not publically available. We have "commented out" the lines in the .do files that create these results, and do not include NELS files in this folder.
Some notes and code (and files) that describe how we go from “raw” NARA FOSF files to the head start #s used in the .zip file above can be found in the following folder:
Code for “Robust inference with Multi-way Clustering”
This .zip file contains a folder with Stata code behind the tables in: Colin Cameron, Jonah Gelbach, and Douglas L Miller, "Robust Inference with Multi-way Clustering", Journal of Business and Economic Statistics, 2011. Please see the readme.txt file for description of the .do files.
Data and Code for “Rearranging the Family”
The .zip file contains the .do files used to extract the data and create the Tables and Figures in Edmonds, Mammen, and Miller “Rearranging the Family? Income Support and Elderly Living Arrangements in a Low-income country”, Journal of Human Resources, 2005. The Stata .dta file contains the “analysis data” used for the paper. (These files were made for posting in October 2014, so there is a chance that I made some error in what is posted. If you think that this is the case, please let me know!)
Rearranging the Family Code (zip file)
Rearranging the Family Data (zip file, 18mb)
Code for “The effects of housing and neighborhood conditions on child mortality”
The .zip file contains the main estimation code for Jacob Ludwig and Miller, “The effects of housing and neighborhood conditions on child mortality”, Journal of Health Economics. (These files were made for posting in October 2014, so there is a chance that I made some error in what is posted. If you think that this is the case, please let me know!)
This page last updated on April 27, 2007.