* Heteroskedasticity in OLS Model for Compensation
*
* Keywords:
*     regression, ols, linear, heteroskedasticity, park, test, compensation
*
* Description:
*     We illustrate how to estimate a linear OLS Model for Compensation
*     and use Park's Heteroskedasticity Test
*
* Author(s):
*     Skif Pankov
*
* Source:
*    Damodar N. Gujarati and Dawn C. Porter, Basic Econometrics - 5th Edition
*    McGraw-Hill International Edition, Chapter 11, Example 11.1 (page 379)
*
sample 1 9

* Reading the datafile and naming variables
read(data_11.1.shd) comp prod

* Running an OLS regression of comp on prod, stating to include residual 
* statistics and to save residuals in a variable reds
ols comp prod / rstat resid = reds

* Requiring to test the previous regression for heteroskedasticity
diagnos / het

* Generating variables
genr lnu2 = log(reds**2)
genr lnprod = log(prod)

* Running an OLS regression of newly generated lnu2 on lnprod, stating to include 
*residual statistics
ols lnu2 lnprod / rstat

stop