* J-Test on OLS Regression for Personal Consumption * * Keywords: * regression, ols, specification, test davidson, mackinnon, j-test, * ppce, per capita, personal, consumption, expenditure * * Description: * We illustrate how to estimate Linear OLS Models for Per Capita Personal * Consumption Expenditure (PPCE) and show how choose between models' * specifications using Davidson-Mackinnon J-Test * * Author(s): * Skif Pankov * * Source: * Damodar N. Gujarati and Dawn C. Porter, Basic Econometrics - 5th Edition * McGraw-Hill International Edition, Chapter 13, Example 13.4 (page 491) * sample 1 36 * Reading the datafile and naming variables read(data_13.4.shd) ppce pdpi * Generating lagged variables genr pdpi1 = lag(pdpi) genr ppce1 = lag(ppce) * Running two different OLS specifications of a per capita consumption model, * stating to start from the second variable (because we are using a lagged * variable) and saving the predicted values of the regression ols ppce pdpi pdpi1 / beg = 2 end = 36 predict = ppcea ols ppce pdpi ppce1 / beg = 2 end = 36 predict = ppceb * Adding predicted values from the first regression onto the second * specification (and vice versa) to conduct the J-test ols ppce pdpi pdpi1 ppceb / beg = 2 end = 36 ols ppce pdpi ppce1 ppcea / beg = 2 end = 36 stop