The ZIP regression is applied for modelling count data with extra zeros. In this paper, we have improved the count parameter estimation in this model with correlated count predictor variables under the assumption of sparseness of the model using Stein-Liu estimators. We have conducted a Monte Carlo simulation study for various combinations of sample sizes, inactive count predictor variables, and correlation level between the count predictor variables in order to compare the performance of the suggested estimator with the unrestricted Liu estimator in terms of simulated relative efficiency criteria. Our results showed that the performance of the Stein-Liu estimators is superior to the unrestricted Liu estimator in any situations.
