Quasi range dependent weighted classes of linear slope estimators

Abstract

The classes of adaptive estimators are proposed to the slope parameter of simple linear regression (SLR) model. These classes are based on the weighted sample slopes where in both slopes and their weights are depending on quasi ranges of predictor variables. The mean, variance of the proposed classes of estimators are obtained and their variances are minimized using numerical method. The performance of some members of these classes are discussed in terms of their relative efficiencies. A simulation study is also carried out to check their behaviour. An illustrative example is provided.