Article ID: | iaor2004439 |
Country: | United States |
Volume: | 21 |
Issue: | 1/2 |
Start Page Number: | 125 |
End Page Number: | 143 |
Publication Date: | Jan 2001 |
Journal: | American Journal of Mathematical and Management Sciences |
Authors: | Corcuera Jos M. |
Keywords: | ARIMA processes |
In this paper the author elucidates how the estimation of the unknown parameters, such as the mean, the variance, or the correlation coefficient, affects the prediction bounds or limits in a first order autoregressive process. Two predictive densities are given, they correct the usual prediction limits in a way that, for very small samples, the new prediction limits are practically exact.