* Log OLS Model for Demand for Chicken
*
* Keywords:
* regression, multiple, log, ols, f-test, demand, chicken
*
* Description:
* We illustrate how to estimate a Multiple Log OLS Regression Model for
* Demand for Chicken and test its coefficients with an F-test
*
* Author(s):
* Skif Pankov
*
* Source:
* Damodar N. Gujarati and Dawn C. Porter, Basic Econometrics - 5th Edition
* McGraw-Hill International Edition, Chapter 8, Example 8.4 (page 253)
*
sample 1 23
* Reading the datafile and naming the variables
read(data_8.4.shd) chkn ipc rpc rpp rpb rps
* Generating logs of variables
genr lnchkn = log(chkn)
genr lnipc = log(ipc)
genr lnrpc = log(rpc)
genr lnrpp = log(rpp)
genr lnrpb = log(rpb)
* Running OLS regression of lnchkn on all other log variables, specifying that
* it is a log-log regression and including residual summary statistics into output
ols lnchkn lnipc lnrpc lnrpp lnrpb / loglog rstat
* Stating the hypothesis that coefficients on lnrpp and lnrpb are both zero and
* testing this hypothesis
test
test lnrpp = 0
test lnrpb = 0
end
stop
* Note:
* The value of F-statistic will be greater that the book value by 0.019 -
* this is because SHAZAM uses more accurate values of sums of squared
* errors