Premium
On Model Selection Based on Validation with Applications to Pressure and Temperature Prognosis
Author(s) -
Hjorth U.,
Holmqvist L.
Publication year - 1981
Publication title -
journal of the royal statistical society: series c (applied statistics)
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.205
H-Index - 72
eISSN - 1467-9876
pISSN - 0035-9254
DOI - 10.2307/2346350
Subject(s) - selection (genetic algorithm) , model selection , model validation , statistics , computer science , mathematics , artificial intelligence , data science
S ummary Model selection by cross‐validation is applied to multivariate autoregressive models for weather data. This is done in a new manner such that the main effects of the model selection are included in the cross‐validation measure. Models selected from different lists of explanatory variables are compared by the method. Examples of competitive short‐time forecasts are given for pressure 6‐24 hr ahead and temperature 6 hr ahead.