Improved parameters estimation in the multicollinear Poisson regression model based on Stein-Liu estimators

نویسندگان

1 Department of Statistics‎, ‎University of Kurdistan‎, ‎Sanandaj‎, ‎Iran

2 Department of Statistics‎, ‎University of Kurdistan‎, ‎Sanandaj‎, ‎Iran

doi
10.22034/jsmta.2025.21347.1134
چکیده

This paper addressed parameter estimation in the Poisson regression model in the presence of multicollinearity when it is surmised that the parameter vector is restricted to a linear subspace‎. ‎To improve the efficiency of parameter estimation‎, ‎we proposed the Stein-Liu and positive Stein-Liu strategies‎. ‎The proposed estimators' asymptotic distributional biases and variances were derived‎, ‎and their variances were compared‎. ‎The performance of the proposed estimators was investigated through an extensive Monte Carlo simulation study‎. ‎The suggested estimators were also applied to data from Swedish football‎. ‎The results confirmed that the performances of our estimators were superior to the unrestricted Liu estimator‎. ‎As an important result‎, ‎the Stein-Liu estimators uniformly perform better than the unrestricted Liu estimator