English

Specification testing in nonparametric AR-ARCH models

Statistics Theory 2016-10-12 v1 Statistics Theory

Abstract

In this paper an autoregressive time series model with conditional heteroscedasticity is considered, where both conditional mean and conditional variance function are modeled nonparametrically. A test for the model assumption of independence of innovations from past time series values is suggested. The test is based on an weighted L2L^2-distance of empirical characteristic functions. The asymptotic distribution under the null hypothesis of independence is derived and consistency against fixed alternatives is shown. A smooth autoregressive residual bootstrap procedure is suggested and its performance is shown in a simulation study.

Keywords

Cite

@article{arxiv.1610.03215,
  title  = {Specification testing in nonparametric AR-ARCH models},
  author = {Marie Hušková and Natalie Neumeyer and Tobias Niebuhr and Leonie Selk},
  journal= {arXiv preprint arXiv:1610.03215},
  year   = {2016}
}