Almost Unbiased Liu Type Estimator in Bell Regression Model: Theory, Simulation and Application

TL;DR

This paper introduces a new almost unbiased Liu-type estimator for the Bell regression model, demonstrating through theory, simulations, and real data that it outperforms existing estimators in mean squared error.

Contribution

The paper proposes a novel almost unbiased Liu-type estimator for the Bell regression model, with theoretical superiority and practical validation.

Findings

Proposed estimator has lower mean squared error than competitors.

Simulation results confirm theoretical advantages.

Real data application shows improved modeling performance.

Abstract

In this paper, we gain the new almost unbiased Liu-type estimators to literature for the Bell regression model. We provide the superiority of the proposed estimator to its competitors such as the maximum likelihood estimator and Liu-type estimators via some theorems. We also design an extensive Monte Carlo simulation study to show that the proposed estimators outperforms the competitors in terms of mean squared error theoretically. Finally, we present a real data study to assess the performance of the introduced estimators in modeling real-life data. The findings of both the simulation and the empirical study demonstrate that the proposed regression estimators surpasses its competitors based on the mean square error criterion.