Recently, the Bell regression model (BRM) is proposed to model a count variable. The BRM is generally preferred over the Poisson regression model to overcome the restriction that the mean is equal to the variance. The BRM is usually estimated using the maximum likelihood estimator (MLE). It is a well-known phenomenon that the MLE is very sensitive to multicollinearity. We propose a Bell Liu regression (BLR) estimator to circumvent the problem of multicollinearity associated with the BRM. Moreover, some new Liu parameters are proposed for the BLR estimator. To evaluate the performance of the proposed estimators, we conduct a Monte Carlo simulation study where the mean squared error is considered as an evaluation criterion. In addition, a real application is also included to show the superiority of the proposed method.

最近，提出了贝尔回归模型 (BRM) 来对计数变量进行建模。BRM 通常优于泊松回归模型，以克服均值等于方差的限制。BRM 通常使用最大似然估计器 (MLE) 进行估计。众所周知，MLE 对多重共线性非常敏感。我们提出了一个 Bell Liu 回归 (BLR) 估计器来规避与 BRM 相关的多重共线性问题。此外，还为 BLR 估计器提出了一些新的 Liu 参数。为了评估所提出的估计器的性能，我们进行了蒙特卡罗模拟研究，其中均方误差被视为评估标准。此外，还包括一个实际应用，以显示所提出方法的优越性。