We consider the problem of testing subhypotheses in a heteroscedastic linear regression model. The proposed test statistics are based on the ranks of scaled residuals obtained under the null hypothesis. Any estimator that is n 1 2 -consistent under the null hypothesis can be used to form the residuals. The error variances are estimated through a parametric model. This extends the theory of aligned rank tests to the heteroscedastic linear model. A real data set is used to illustrate the procedure.
All Science Journal Classification (ASJC) codes
- Statistics and Probability
- Statistics, Probability and Uncertainty
- Applied Mathematics