{"id":565,"date":"2017-09-28T03:43:55","date_gmt":"2017-09-28T03:43:55","guid":{"rendered":"http:\/\/jsr.isrt.ac.bd\/?post_type=article&p=565"},"modified":"2017-09-28T03:44:15","modified_gmt":"2017-09-28T03:44:15","slug":"class-estimators-regression-coefficient-sign-change-problem-measurement-errors","status":"publish","type":"article","link":"http:\/\/jsr.isrt.ac.bd\/article\/class-estimators-regression-coefficient-sign-change-problem-measurement-errors\/","title":{"rendered":"A class of estimators of regression coefficient for sign change problem in measurement errors"},"content":{"rendered":"
Srivastava and Shalabh (1997a, Journal of Econometrics) proposed a class of
\nStein-like consistent estimators for estimating the slope coefficient in a single explanatory
\nvariable ultrastructural model. This paper studies the sign reversal
\nproblem of this class of estimators and proposes an alternative class of improved
\nestimators along the lines of the double- class estimators of Ullah and Ullah
\n(1978,\\textit{ Econometrica}) that overcomes this problem. Large sample asymptotic
\nproperties of the proposed estimators are studied for the case where the distributions
\nof the measurement errors are not necessarily normal.<\/p>\n
<\/p>\n