Title
Detections of changes in return by a wavelet smoother with conditional heteroscedastic volatility
Abbreviated Journal Title
J. Econom.
Keywords
nonparametric regression; wavelet coefficient; change points; kernel; estimation; local polynomial smoother; conditional heteroscedastic; variance; alpha-mixing; CHANGE-POINTS; NONPARAMETRIC REGRESSION; TERM STRUCTURE; TIME-SERIES; MODELS; JUMP; IDENTIFICATION; DIFFUSION; VARIANCE; Economics; Mathematics, Interdisciplinary Applications; Social Sciences, ; Mathematical Methods
Abstract
In this paper, we propose two estimators, an integral estimator and a discretized estimator, for the wavelet coefficient of regression functions in nonparametric regression models with heteroscedastic variance. These estimators can be used to test the jumps of the regression function. The model allows for lagged-dependent variables and other mixing regressors. The asymptotic distributions of the statistics are established, and the asymptotic critical values are analytically obtained from the asymptotic distribution. We also use the test to determine consistent estimators for the locations of change points. The jump sizes and locations of change points can be consistently estimated using wavelet coefficients, and the convergency rates of these estimators are derived. We perform some Monte Carlo simulations to check the powers and sizes of the test statistics. Finally, we give practical examples in finance and economics to detect changes in stock returns and short-term interest rates using the empirical wavelet method. (C) 2007 Elsevier B.V. All rights reserved.
Journal Title
Journal of Econometrics
Volume
143
Issue/Number
2
Publication Date
1-1-2008
Document Type
Article
Language
English
First Page
227
Last Page
262
WOS Identifier
ISSN
0304-4076
Recommended Citation
"Detections of changes in return by a wavelet smoother with conditional heteroscedastic volatility" (2008). Faculty Bibliography 2000s. 203.
https://stars.library.ucf.edu/facultybib2000/203
Comments
Authors: contact us about adding a copy of your work at STARS@ucf.edu