Abstrakt
There exist a number of tests for assessing the nonparametric heteroskedastic location scale assumption. The goodness-of-fit tests considered are for the more general hypothesis of the validity of this model under a parametric functional transformation on the response variable, specifically testing for independence between the regressors and the errors in a model where the transformed response is just a location/scale shift of the error is considered.
The proposed criteria use the familiar factorization property of the joint characteristic function under independence. The difficulty is that the errors are unobserved and hence one needs to employ properly estimated residuals in their place.
The limit distribution of the test statistics is studied under the null hypothesis as well as under alternatives. and also a resampling procedure is suggested in order to approximate the critical values of the tests. This resampling is subsequently employed in a series of Monte Carlo experiments that illustrate the finite-sample properties of the new test.
The performance of related test statistics for normality and symmetry of errors is also investigated, and application of our methods on real data sets is provided. (C) 2019 Elsevier B.V. All rights reserved.