A Comparative Study of Test Procedures uses in Assessing the Forecasting Ability of Linear Models with Applications to Crop Yield Data
The choice of the appropriate linear model before this can be used for planning and decision making, has been the concern of many statistical workers. Most of the methods in the literature aim at evaluating the descriptive ability of the candidate models. In the present paper an evaluation scheme of the predictability of a linear model based on a function of the discrepancy of the observed and the corresponding predicted values of the dependent variable is studied. Based on this statistical function, the predictability of a linear model is tested. Considering the ratio of such functions for two linear models, the predictability of these models is compared. Applications on real and simulated data are also presented
|Date of creation:||1998|
|Date of revision:|
|Publication status:||Published in 4th Hellenic European Conference on Computer Mathematics and its Applications (1998): pp. 501-508|
|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:6280. 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.