Abstract:
It is well known that the use of prior information in the logistic regression improves
the estimates of regression coefficients when multicollinearity presents. This prior
information may be in the form of exact or stochastic linear restrictions. In this article,
in the presence of stochastic linear restrictions, we propose a new efficient estimator,
named Stochastic restricted optimal logistic estimator for the parameters in the logistic
regression models when the multicollinearity presents. Further, conditions for the
superiority of the new optimal estimator over some existing estimators are derived with
respect to the mean square error matrix sense. Moreover, a Monte Carlo simulation
study and a real data example are provided to illustrate the performance of the proposed
optimal estimator in the scalar mean square error sense.
