On a Testing Procedure for Model Selection
In this paper a forecasting model selection scheme is considered which amounts to testing the predictive behaviour of a model by adopting Xekalaki and Katti's (1984) idea of assigning to its performance a score for each of a series of time points. The score reflects how close to, or how far from, the predictive value the observed actual value is. A statistical test is proposed for comparing the forecasting performances of two models
|Date of creation:||1991|
|Date of revision:|
|Publication status:||Published in Proceedings of the 48th Session of the International Statistical Institute, Cairo (1991): pp. 513-514|
|Contact details of provider:|| Postal: Ludwigstraße 33, D-80539 Munich, Germany|
Web page: https://mpra.ub.uni-muenchen.de
More information through EDIRC
When requesting a correction, please mention this item's handle: RePEc:pra:mprapa:6260. See general information about how to correct material in RePEc.
For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: (Joachim Winter)
If references are entirely missing, you can add them using this form.