The researcher suspects that the original regression errors might exhibit ARCH-t
ID: 3270024 • Letter: T
Question
The researcher suspects that the original regression errors might exhibit ARCH-type heteroskedasticity thus she runs the regression below, where the variable RESID contains the residuals from the regression above, the variable RESID^2 contains the squared residuals, and the variable RESID^(-1) contains the squared residuals at time t-1. Use these results to test for heteroskedasticity, stating very clearly the null and alternative hypotheses for your test, the auxiliary regression, the test statistic and its distribution under the null.Explanation / Answer
ho: the assumption of homogenity of variance is valid
h1: the assumption of homogenity of variance is NOT valid
with F = 22.8 and p-value < .05 i reject ho and conclude that the assumption of homogenity of variance is NOT valid