Back to Departmental Colloquium: Spring 2015
Departmental Colloquium
Date: Thursday, Feb 26, 2015
Time: 4:00PM
Location: LCB 219
Alexander Aue
University of California, Davis
Title |
Piecewise quantile autoregressive modeling of non-stationary time series |
Abstract |
A new methodology is discussed for the fitting of non-stationary time series that exhibit non-linearity, asymmetry, local persistence and changes in location, scale and shape of the underlying distribution. To do this, model selection techniques are developed for the class of piecewise stationary quantile autoregressive processes. The best model is defined in terms of minimizing a minimum description length criterion derived from an asymmetric Laplace likelihood. Its practical minimization is achieved with the use of genetic algorithms. If the data generating process follows indeed a piecewise quantile autoregressive structure, it can be shown that the proposed method is consistent for estimating the number of pieces and the autoregressive parameters. Empirical work suggests that the proposed method performs well in finite samples. The talk is based on joint work with R. Cheung, T. Lee and M. Zhong. |
Mathematical Biology
Computational Mathematics
Applied Mathematics