© 2003 by Biometrika Trust
Testing and estimation of thresholds based on wavelets in heteroscedastic threshold autoregressive models
1 Department of Applied Mathematics, The Hong Kong Polytechnic University, Kowloon, Hong Kongmathipwc{at}polyu.edu.hk mathwong{at}polyu.edu.hk 2 Department of Mathematics, Guangzhou University, Guangzhou 510405, China.yli57{at}hotmail.com 3 Institute of Applied Mathematics, The Chinese Academy of Sciences, Beijing 100080, China. ahz{at}mail.amt.ac.cn
We consider the testing and estimation of thresholds in heteroscedastic threshold autoregressive models with an unknown number of thresholds. A test statistic based on empirical wavelet coefficients is proposed.The asymptotic distribution of the test statistic is established and consistent estimators of the thresholds and the number of thresholds are given. A Monte Carlo study and a real example are used to assess the performance of our method.
Key Words: Heteroscedasticity; Jump point; Threshold; Time delay; Wavelet
Received September 2001. Revised October 2002