A0589
Title: Goodness-of-fit tests in functional-coefficient autoregressive models with measurement error
Authors: Pei Geng - University of New Hampshire (United States) [presenting]
Abstract: The functional-coefficient autoregressive (FAR) models are flexible to fit the nonlinear patterns in time series data. The aim is to introduce a goodness-of-fit test for the FAR models when the time series is observed with measurement error. The calibrated autoregressive model based on the observed time series is represented, and the test for the parametric functional coefficients is constructed based on a marked empirical process with the residuals and the covariate. The asymptotic property shows that the proposed test is asymptotically distribution-free. A finite-sample simulation study is conducted to demonstrate the empirical level and power under certain alternatives.
