A1575
Title: A nonparametric test for change-points in volatility with weighted empirical processes
Authors: Ba Chu - Carleton University (Canada) [presenting]
Abstract: Financial markets have witnessed significant episodes of structural instability, which can lead to changes in volatility and possibly other risk measures. A new test is proposed for changes in volatility in a nonparametric heteroscedastic regression model, $Y_t = \mu[{\mathbf{X}_t}] + \sigma[{\mathbf{X}_t}]\epsilon_t,$ where $\epsilon_t$ is a strictly stationary sequence of errors. To achieve this goal, the limiting behaviour of weighted empirical processes is first studied with weakly dependent and stationary data. A bootstrap procedure is also proposed to improve the finite-sample performance of the proposed approach.