Confidence regions for parameters in linear models with additional information |
| |
Authors: | K Henschke |
| |
Institution: | Sektion Mathetsmatik , Humboldt-Universit?t , Unter den Linden 6 Postfach 1297, Berlin, DDR-1086 |
| |
Abstract: | Using given significant additional information it is possible to improve different confidence regions for the regression parameters in a linear model. Thereby, the given informations may concern the expectation and (or) the variance of the observations, and an improvement is possible in the sense of the decrease of the confidence regions' size. In particular it is possible to improve the so called confidence ellipsoids which are often used to estimate the considered parameters. |
| |
Keywords: | Confidence regions estimation of regression parameters linear model convex model additional information improvement of confidence region? |
|
|