A distribution-free transform of the residuals sample autocorrelations with application to model checking
AbstractWe propose an asymptotically distribution-free transform of the sample autocorrelations of residuals in general parametric time series models, possibly non-linear in variables. The residuals autocorrelation function is the basic model checking tool in time series analysis, but it is useless when its distribution is incorrectly approximated because the effects of parameter estimation or of unnoticed higher order serial dependence have not been taken into account. The limiting distribution of residuals sample autocorrelations may be difficult to derive, particularly when the underlying innovations are not independent. However, the transformation we propose is easy to implement and the resulting transformed sample autocorrelations are asymptotically distributed as independent standard normals, providing an useful and intuitive device for model checking by taking over the role of the standard sample autocorrelations. We also discuss in detail alternatives to the classical Box-Pierce and Bartlett's Tp-process tests, showing that our transform entails no efficiency loss under Gaussianity. The finite sample performance of the procedures is examined in the context of a Monte Carlo experiment for the two goodness-of-fit tests discussed in the article. The proposed methodology is applied to modeling the autocovariance structure of the well known chemical process temperature reading data already used for the illustration of other statistical procedures.
Download InfoIf you experience problems downloading a file, check if you have the proper application to view it first. In case of further problems read the IDEAS help page. Note that these files are not on the IDEAS site. Please be patient as the files may be large.
Bibliographic InfoPaper provided by Universidad Carlos III, Departamento de Economía in its series Economics Working Papers with number we101707.
Date of creation: Apr 2010
Date of revision:
Residuals autocorrelation function; Asymptotically pivotal statistics; Nonlinear in variables models; Long memory; Higher order serial dependence; Recursive residuals; Model checking; Local alternatives;
This paper has been announced in the following NEP Reports:
- NEP-ALL-2010-04-24 (All new papers)
- NEP-ECM-2010-04-24 (Econometrics)
- NEP-ETS-2010-04-24 (Econometric Time Series)
Please report citation or reference errors to , or , if you are the registered author of the cited work, log in to your RePEc Author Service profile, click on "citations" and make appropriate adjustments.:
- Francq, Christian & Roy, Roch & Zakoian, Jean-Michel, 2005. "Diagnostic Checking in ARMA Models With Uncorrelated Errors," Journal of the American Statistical Association, American Statistical Association, vol. 100, pages 532-544, June.
- Hong, Yongmiao, 1996. "Consistent Testing for Serial Correlation of Unknown Form," Econometrica, Econometric Society, vol. 64(4), pages 837-64, July.
- Lobato, I.N. & Nankervis, John C. & Savin, N.E., 2002. "Testing For Zero Autocorrelation In The Presence Of Statistical Dependence," Econometric Theory, Cambridge University Press, vol. 18(03), pages 730-743, June.
- Miguel A. Delgado & Javier Hidalgo & Carlos Velasco, 2005. "Distribution Free Goodness-of-Fit Tests for Linear Processes," STICERD - Econometrics Paper Series /2005/482, Suntory and Toyota International Centres for Economics and Related Disciplines, LSE.
- Durbin, J, 1970. "Testing for Serial Correlation in Least-Squares Regression When Some of the Regressors are Lagged Dependent Variables," Econometrica, Econometric Society, vol. 38(3), pages 410-21, May.
For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: ().
If references are entirely missing, you can add them using this form.