Abstract Details
Activity Number:
|
245
|
Type:
|
Contributed
|
Date/Time:
|
Monday, August 4, 2014 : 2:00 PM to 2:45 PM
|
Sponsor:
|
Section for Statistical Programmers and Analysts
|
Abstract #314042
|
|
Title:
|
A New Goodness-of-Fit Test for Time Series Models Based on Correlation Between the Sample Autocorrelation and Partial Autocorrelation Sequences
|
Author(s):
|
James Faulkner*+ and Donald B. Percival
|
Companies:
|
University of Washington and University of Washington
|
Keywords:
|
time series ;
goodness-of-fit ;
power ;
simulation
|
Abstract:
|
An important step in time series analysis is proper description of the covariance structure of the process under study. Residuals from a model that correctly specifies the underlying true process should obey a white noise process. We propose a new goodness-of-fit test based upon a previously undocumented correlation between the sample autocorrelation and partial autocorrelation sequences that approaches unity at small lags under the null hypothesis of white noise. We compare the power of the proposed test to several other known portmanteau tests using data simulated from autoregressive moving average (ARMA) and fractionally differenced processes. We show that the proposed test is as or more powerful than the other tests for most simulated processes, with the greatest improvements in power realized for small to moderate sample sizes and moderate autocorrelations. Our new test is computationally simple and does away with the arbitrary selection of lag orders necessary with other portmanteau tests, which will make it a useful addition to the current set of time series goodness-of-fit tests.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2014 program
|
2014 JSM Online Program Home
For information, contact jsm@amstat.org or phone (888) 231-3473.
If you have questions about the Professional Development program, please contact the Education Department.
The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.
Copyright © American Statistical Association.