Activity Number:
|
294
|
Type:
|
Contributed
|
Date/Time:
|
Tuesday, July 31, 2007 : 10:30 AM to 12:20 PM
|
Sponsor:
|
Business and Economics Statistics Section
|
Abstract - #308254 |
Title:
|
A New Type of Portmanteau Test Based on the Discrete Cosine Transform
|
Author(s):
|
Sungun Oh*+ and Hyemin Cho and In-Kwon Yeo
|
Companies:
|
Sookmyung Women's University and Sookmyung Women's University and Sookmyung Women's University
|
Address:
|
Department of Statistics, Seoul, 140-742, South Korea
|
Keywords:
|
Discrete cosine transform ; Multivariate portmanteau test ; Ljung-Box test ; Determinant of covariance
|
Abstract:
|
We present a new type of portmanteau test in the frequency domain which is derived from the discrete cosine transform (DCT). For the stationary time series, DCT coefficients are asymptotically independent and their variances are expressed by linear combinations of autocovariances. The covariance matrix of DCT coefficients for white noises is diagonal matrix whose diagonal elements is the variance of time series. A simple way to test the independence of time series is that we divide DCT coefficients into two or more sets and then compare their sample variances. We also do this by testing the slope in the linear regression model of which the response variables are absolute values or squares of coefficients. Similar approaches are applied to multivariate time series. Simulation results show that the proposed tests have much higher powers than existing tests in most cases of our experiments.
|