Linearized shrinkage estimator for Semiparametric Regression Model

Abstract

Semiparametric regression models are an appropriate tool to study complex data sets since they make use of elements of parametric and nonparametric regression. The nonparametric methods are effective when there are variables that operate by a particular rule, and others that have their effects that are more intricate or hard to predict. With the help of semiparametric regression model, researchers run the risk of revealing the problem of multicollinearity. This is the reason why it is important to consider the shrinkage parameters when analyzing the Liu estimator data. Within the Liu estimator, the parameter of shrinkage is set according to several steps which are discussed and described. After this, the slime mould algorithm (SMA) has been applied in selecting the appropriate parameter of shrinkage in Liu regression when dealing with issues of multicollinearity. In our study, it was found that there are estimators that would significantly improve on mean squared error when compared to their counterparts.