# st: Hausman& Taylor

 From "Antonio Rodrigues Andres" To Subject st: Hausman& Taylor Date Sun, 28 Dec 2003 19:23:41 +0100

```Dear Stata users

I have written this STATA code to implement the Hausman and Taylor
estimator but I am not quite sure if this is the right procedure, in
particular, to get consistent estimates of the variance components.

Someone could help me

ALl the best

Antonio

>>>>>>>>>>>>>

REPLICATE PAGES 120-121: HAUSMAN TAYLOR APPROACH USING WAGES DATA

*GENERATE LIST OF ENDOGENOUS VARIABLE;

global yvar "lwage";

* GENERATE LIST OF EXPLANATORY VARIABLES (Xit and Zi);
global z1 "fem blk"
global z2 "ed"
global x1 "wks south smsa ms"
global x2 "occ ind union exp exp2"

*WITHIN REGRESSION;
regress devlwage devwks devsouth devsmsa devms devocc devind devunion
devexp2 devmy76 devmy77 devmy78 devmy79 devmy80 devmy81 devmy82,
noconstant

*WITHIN RESIDUALS;
predict residu, residuals
ge sqresidu=residu*residu

*SQUARED SUM OF WITHIN RESIDUALS;
egen sumsqru=sum(sqresidu)

*CREATE SIGMA2ETA (Hausman Taylor p. 1384). consistent estimate of
sigma2eta;
gen sigmau=sumsqru/(4164-595) */ NT=4164 N=595
su sumsqru sqresidu sigmau

*GENERATE di (group means of within residuals);
egen mresidu=mean(residu), by (id)

*2SLS of di on z1 and z2 using as instruments x1 and z1;
ivreg mresidu fem blk (ed= wks south smsa ms)

*save residuals
predict resiv, residuals
gen sqresiv=resiv*resiv

* squared sum of 2SLS residuals
egen sumresiv=sum(sqresiv)

* consistent estimate sigma2 (1/n)*/(yi.-xi.*bw-zi*gammaw)'
(yi.-xi.*bw-zi*gammaw), p. 1384 HT where yi. =vector of group means,
bw=within estimate gammav=2SLS estimates

*gammav is obtained from the above 2SLS regression
gen sigmat=sumresiv/595
>>>>>>>>>>>>>>>>>>>>>>

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/
```