We illustrate the roles of three types of diagnostics for
detecting or confirming the presence of trading day effects and for
selecting among competing trading day models. These are spectral
analysis, AIC comparisons, and out-of-sample forecast-error
diagnostics. We show merits and limitations of each in the context
of a study of four alternative trading day models for fifty-eight
Census Bureau time series. We consider both direct estimation from
the observations and indirect estimation from the irregulars.
Spectral analysis, regARIMA models, model selection, forcasting,
monthly time series
This symbol indicates a link to a non-government web site. Our linking to these sites does not constitute an endorsement of any products, services or the information found on them. Once you link to another site you are subject to the policies of the new site.