{"id":518,"date":"2017-09-24T06:31:46","date_gmt":"2017-09-24T06:31:46","guid":{"rendered":"http:\/\/jsr.isrt.ac.bd\/?post_type=article&p=518"},"modified":"2017-09-24T06:32:29","modified_gmt":"2017-09-24T06:32:29","slug":"stein-rule-estimation-ultrastructural-model-exact-linear-restrictions","status":"publish","type":"article","link":"http:\/\/jsr.isrt.ac.bd\/article\/stein-rule-estimation-ultrastructural-model-exact-linear-restrictions\/","title":{"rendered":"Stein-rule estimation in ultrastructural model under exact linear restrictions"},"content":{"rendered":"
The role and construction of Stein-rule estimators in multivariate ultrastructural
\nmodel is discussed when some prior information about the regression coe\u000ecients
\nis available in the form of exact linear restrictions. The additional information
\nin the forms of covariance matrix of measurement errors and reliability matrix
\nof explanatory variables is used for the construction of consistent estimators.
\nTwo families of Stein-rule estimators are proposed using each type of additional
\ninformation which are consistent as well as satisfy the exact linear restrictions.
\nThe distribution of measurement errors is assumed to be not necessarily normally
\ndistributed. The asymptotic distribution of the proposed families of Stein-rule
\nestimators are derived and studied. The \fnite sample properties of the estimators
\nare studied through a Monte-Carlo simulation experiment.<\/p>\n