Robust biased estimators for Poisson regression model: Simulation and applications

The method of maximum likelihood flops when there is linear dependency (multicollinearity) and outlier in the generalized models. In this study, we combined ridge estimator with transformed M-estimator (MT) conditionally unbiased bounded influence (CE). two new estimators are called robust MT Robust-CE. A Monte Carlo study revealed that proposed dominate for models Poisson response log link function. real-life application results support simulation outcome.