DoctorateOpen Access

Inadmissibility of the stein-rule estimators under the loss functions

2006
0 views
0 downloads
Advisor: Prof.dr. Fikri Akdeniz

Abstract (EN)

In applied fields of Statistics, such as econometrics or medical sciences, especially ingeneral linear model, it?s commonly seen that the number of parameters to be estimatedis greater than 2. In such situations, estimating the parameters by ordinary least squaresestimation method does not give the correct results. To cope with this problem, Stein (1956)proposed Stein-rule estimators giving better results than ordinary least squares estimationmethod, under the criteria of squared error loss function, if the number of parameters isgreater than 2.In this study, first of all, the properties of Stein-rule estimators will be given, then somefundamental information will be given about these estimators by analayzing the previousworks in this field. In the final chapter, using these information, the recently found resultsabout Stein-rule estimators will be presented.Key Words: Admissibility, Estimators, Loss Functions, Stein-Rule Estimators

Author

Deniz Ünal

How to Cite

Deniz Ünal (Doctorate thesis). Inadmissibility of the stein-rule estimators under the loss functions, 2006, Çukurova University.

Keywords

License

Tüm Hakları Saklıdır

This work is shared under the specified license terms.

More theses from Çukurova University