الفهرس 
Statistical inference of constrained principal components models
AbstractOnly 14 pages are availabe for public view
 |  | from | 147 |  |  |
| | | from | 147 | | |
Abstract
A serious problem in the statistical estimation has been found if a false hypothesis has not been rejected because of ignoring prior information about the estimator. The constrained principal component model is a general model for many constrained estimator. This model ignores the variance of the error term found in the restriction model. Another face for the constrained principal component model has been found. Its variance of the error term that found in the restriction model has a non-zero value. A generalized ordinary mixed estimator (GOME) using the constrained principal component model which show by Takane (2014) has been introduced. Some special cases for this estimator have been found. The new estimator is more benefit for researchers in making decisions and depending on results that have more credibility. The subset models and its constrained models have been got from the constrained principal component model. A model and a constrained model each of them have a variance term that has a non-zero value. A mixed estimator for each model has been got and combined them to get the generalized ordinary mixed estimator. The Superiority of the new estimator has been found to support it