Title
Testing the stability of the functional autoregressive process
Abstract
The functional autoregressive process has become a useful tool in the analysis of functional time series data. It is defined by the equation X"n"+"1=@JX"n+@e"n"+"1, in which the observations X"n and errors @e"n are curves, and @J is an operator. To ensure meaningful inference and prediction based on this model, it is important to verify that the operator @J does not change with time. We propose a method for testing the constancy of @J against a change-point alternative which uses the functional principal component analysis. The test statistic is constructed to have a well-known asymptotic distribution, but the asymptotic justification of the procedure is very delicate. We develop a new truncation approach which together with Mensov's inequality can be used in other problems of functional time series analysis. The estimation of the principal components introduces asymptotically non-negligible terms, which however cancel because of the special form of our test statistic (CUSUM type). The test is implemented using the R package fda, and its finite sample performance is examined by application to credit card transaction data.
Year
DOI
Venue
2010
10.1016/j.jmva.2008.12.008
J. Multivariate Analysis
Keywords
Field
DocType
asymptotic justification,observations x,equation x,functional principal component analysis,test statistic,credit card transaction data,principal component,functional autoregressive process,functional time series data,change-point,functional time series analysis,62m10,time series data,autoregressive process,transaction data,asymptotic distribution,time series analysis
Functional principal component analysis,Econometrics,Truncation,Time series,Autoregressive model,Test statistic,Credit card,Statistics,Statistical hypothesis testing,Mathematics,Asymptotic distribution
Journal
Volume
Issue
ISSN
101
2
Journal of Multivariate Analysis
Citations 
PageRank 
References 
4
1.79
2
Authors
3
Name
Order
Citations
PageRank
Lajos Horváth1289.10
Marie Hušková2239.73
Piotr Kokoszka3278.46