论文信息 - On Model Selection Based on Validation with Applications to Pressure and Temperature Prognosis

On Model Selection Based on Validation with Applications to Pressure and Temperature Prognosis

SUMMARY 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.

U. Hjorth | L. Holmqvist | U. Hjorth | L. Holmqvist

[1] H. Akaike. Statistical predictor identification , 1970 .

[2] N. Draper,et al. Applied Regression Analysis , 1966 .

[3] J. Holton. An introduction to dynamic meteorology , 2004 .

[4] Seymour Geisser,et al. The Predictive Sample Reuse Method with Applications , 1975 .

[5] U. Hjorth. Model Selection and Forward Validation , 1982 .

[6] N. Draper,et al. Applied Regression Analysis. , 1967 .

[7] R. Shibata. Asymptotically Efficient Selection of the Order of the Model for Estimating Parameters of a Linear Process , 1980 .

[8] M. Stone. Cross‐Validatory Choice and Assessment of Statistical Predictions , 1976 .

[9] R. R. Hocking. The analysis and selection of variables in linear regression , 1976 .