| Title: |
Approximate Confidence Regions for Minimax-Linear Estimators |
| Authors: |
Minimax-linear Estimators; H. Toutenburg; A. Fieger; B. Schaffrin |
| Contributors: |
The Pennsylvania State University CiteSeerX Archives |
| Source: |
ftp://ftp.stat.uni-muenchen.de/pub/sfb386/paper166.ps.Z |
| Publication Year: |
1999 |
| Collection: |
CiteSeerX |
| Description: |
Minimax estimation is based on the idea, that the quadratic risk function for the estimate fi is not minimized over the entire parameter space IR K , but only over an area B(fi) that is restricted by a priori knowledge. If all restrictions define a convex area, this area can often be enclosed in an ellipsoid of the form B(fi) = ffi : fi 0 T fi rg. The ellipsoid has a larger volume than the cuboid. Hence, the transition to an ellipsoid as a priori information represents a weakening, but comes with an easier mathematical handling. Deriving the linear Minimax estimator we see that it is biased and nonoperationable. Using an approximation of the non-central 2 -distribution and prior information on the variance, we get an operationable solution which is compared with OLSE with respect to the size of the corresponding confidence intervals. 1 Introduction We consider the linear regression model y = Xfi + ffl; ffl N(0; oe 2 I) (1) with nonstochastic regressor matrix X of full co. |
| Document Type: |
text |
| File Description: |
application/postscript |
| Language: |
English |
| Relation: |
http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.42.4663 |
| Availability: |
http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.42.4663 |
| Rights: |
Metadata may be used without restrictions as long as the oai identifier remains attached to it. |
| Accession Number: |
edsbas.D3B44B31 |
| Database: |
BASE |