Activity Number:
|
503
- SPAAC Poster Competition
|
Type:
|
Topic Contributed
|
Date/Time:
|
Wednesday, August 2, 2017 : 10:30 AM to 12:20 PM
|
Sponsor:
|
Scientific and Public Affairs Advisory Committee
|
Abstract #323500
|
|
Title:
|
R/S-Bootstrapping Test for Fractional Integration
|
Author(s):
|
Yixun Xing* and Wayne A. Woodward
|
Companies:
|
Southern Methodist University and Southern Methodist University
|
Keywords:
|
Long memory ;
R/S-type statistic ;
Bootstrap ;
Time series
|
Abstract:
|
The autocorrelation of long memory processes decays much slower than that of short memory processes, e.g. autoregressive processes. Fractional integration (FI) is a basic model that shows long memory behavior. To detect the FI processes, the R/S-type statistic has been designed by Hurst (1951) and Lo (1991) and used by Giraitis et al. (2003). However, the power tends to be insufficient when the size is controlled. The authors investigate a parametric bootstrap procedure proposed by Woodward et al. (1998) based on the R/S-type statistics for testing for long memory of the FI type. The simulation results show improvements in size and power compared with tests used by Giraitis et al. (2003) which were based on the asymptotic distribution of the R/S-type statistics.
|
Authors who are presenting talks have a * after their name.