* Multicollinearity in the Longley Data * * Keywords: * regression, ols, multicollinearity, longley * * Description: * We illustrate the problems which Multicollinearity can bring by using * Longley dataset (notorious for severe multicollinearity) and estimating * the same model on datasets different only in one observation * * Author(s): * Noel Roy * Skif Pankov * * Source: * William H. Greene, Econometric Analysis - 7th Edition * Pearson International Edition, Chapter 4, Example 4.11 (page 130) * * Setting the first time period to be equal to year 1947 with periodicity of * one year time 1947 1 sample 1947.0 1962.0 * Reading the datafile and naming the variables, specifying to ignore the * first line of the file read (TableF4-2.shd) employ gnpdef gnp armedfor year / skiplines=1 * Running OLS regressions of employ on year, gnpdef, gnp and armedfor, * specifying to use the dataset up to the observation from 1961 (exclude the * last, 16th observation) in the first regression ols employ year gnpdef gnp armedfor / END=1961.0 ols employ year gnpdef gnp armedfor * Printing the correlation matrix of variables ?stat employ year gnpdef gnp armedfor / pcor * Printing the r**2 statistics for the auxiliary regressions of each independent * variable on all other independent variables. Variance inflation = 1/(1-r**2) ?ols employ year gnpdef gnp armedfor / auxrsqr stop