Comparison of the kernel regression and spline regression methods in estimating the Poisson regression model with a practical application

Abstract

The Poisson Dist. belongs to the intermittent mathematical distributions, which are very important in many different phenomena, and rare phenomena such as train rollovers, suicides, etc., and there are many mathematical models, the most prominent of which are regression models, specifically the equations converted from statistical distributions that are used for prediction and estimation, As a result of the variation in the quality of the estimated regression models and the invalidity of using some of the models because they do not have the characteristics of good estimators, it leads to a lack of confidence in their predictive or estimating accuracy, which necessitated the estimation of the generalized Poisson regression equation using nonparametric methods, which are the two methods of nucleus regression and slide regression in modeling the relationship between the number of times the patient was exposed to heart attacks, the response variable (Y) and the patient's age, the exudative variable (X) for a sample size (70) in Dhi Qar governorate, and it was shown through the results that It was concluded that the slide regression method is the best because it has the lowest MSE, and that the value of the pseudo interpretation coefficient R-square was (0.8206)