{"id":627,"date":"2017-09-28T06:33:05","date_gmt":"2017-09-28T06:33:05","guid":{"rendered":"http:\/\/jsr.isrt.ac.bd\/?post_type=article&p=627"},"modified":"2017-09-28T06:33:23","modified_gmt":"2017-09-28T06:33:23","slug":"small-sample-properties-improved-estimators-logistic-regression-skew-normally-distributed-explanatory-variables","status":"publish","type":"article","link":"http:\/\/jsr.isrt.ac.bd\/article\/small-sample-properties-improved-estimators-logistic-regression-skew-normally-distributed-explanatory-variables\/","title":{"rendered":"Small-sample properties of some improved estimators in logistic regression with skew-normally distributed explanatory variables"},"content":{"rendered":"
This study explores the small-sample properties of five estimators (the unrestricted
\nmaximum likelihood estimator, the shrinkage restricted estimator, the
\nshrinkage preliminary test estimator, the shrinkage estimator and the positiverule
\nshrinkage estimator) using Monte Carlo experiments to confirm the asymptotic
\nfindings of Matin and Saleh (2005). It also explores the properties of test
\nprocedures (the Wald, the score and the likelihood ratio) in performing in estimators
\nand tests under consideration. This study confirms the theoretical results
\nin cases where comparisons are possible. When the number of explanatory variables
\nis greater than or equal to 3 the shrinkage and the positive-rule shrinkage
\nestimators always perform well. Considering the MSE the positive-rule shrinkage
\nestimator performs better than the shrinkage estimator. The likelihood ratio test
\nstands out to be the best. However, we lean toward the use of the Wald statistic
\nwhen the problem of estimation is of paramount interest as it provides lower bias
\nand MSE for the estimators.<\/p>\n