PREDICCION DE SERIES DE VENTAS: UN ANALISIS DE COINTEGRACION CON EL PIB

Pablo Marshall ()
Additional contact information
Pablo Marshall: Escuela de Administración, Pontificia Universidad Católica de Chile

Abante, 1998, vol. 1, issue 1, 89-109

Abstract: This study considers different methodologies for times series prediction of Sales. Given the specification of a bivariate autoregressive time series model for Sales and GDP, the paper derives classical models of prediction, such as a regression model, a regression model with error correction, a transfer function, and an ARIMA model. The study discusses the relevance of these models when the Sales series are cointegrated with GDP and when they are not. Results for 20 quarterly Sales time series between 1983 y 1996, show that GDP does not significantly explain short-term variations in Sales. The long-term variations in sales are completely explained by GDP in 6 out of the 20 series. The best results for short-term out-of-sample predictions are obtained with the ARIMA model. For long-term predictions, if the Sales time series are cointegrated with the GDP, the best is the regression model. Otherwise, the ARIMA model gives better results. The results of this study are a contribution to the selection of the best methodologies for short- and long-term time series prediction of Sales.

Keywords: Series prediction; Cointegrated; Emerging Markets; Chile (search for similar items in EconPapers)
JEL-codes: C13 M10 (search for similar items in EconPapers)
Date: 1998
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
http://www.abante.cl/files/ABT/Contenidos/Vol-1-N1/4.%20Marshall.pdf (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:pch:abante:v:1:y:1998:i:1:p:89-109

Ordering information: This journal article can be ordered from

Access Statistics for this article

Abante is currently edited by Eduardo Walker

More articles in Abante from Escuela de Administracion. Pontificia Universidad Católica de Chile. Contact information at EDIRC.
Bibliographic data for series maintained by Eduardo Walker ().