Forecasting with the maximum of correlated components having bivariate t-distributed errors

Forecasting with the maximum of correlated components having bivariate t-distributed errors

0.00 Avg rating—0 Votes
Article ID: iaor19982007
Country: United Kingdom
Volume: 7
Issue: 3
Start Page Number: 233
End Page Number: 237
Publication Date: Jul 1996
Journal: IMA Journal of Mathematics Applied in Business and Industry
Authors:
Keywords: statistics: regression, forecasting: applications
Abstract:

The value of the maximum of two forecasts as a component in a linear combination forecast is assessed when the errors have a bivariate t distribution. It is found that, while the maximum may provide information additional to that of the mean forecast if the components are not equally biased and have heteroscedastic errors, the best linear combination forecast does not give any weight to the maximum.

Reviews

Required fields are marked *. Your email address will not be published.