Employing the Kernel method in estimating the Poisson regression model with application

Abstract

Statistics has been an extended hand that provides its services over time to other sciences and a tributary that contributes to analyzing the results of studies in various fields and areas of knowledge. The Poisson regression model is one of the vital statistical models to represent natural phenomena, and the (kernel) method is described as one of the most important parametric methods used to estimate its parameters. Therefore, the study included the use of the Poisson regression model in formulating relationships in which the response variable has numerical values ​​with a random error that follows its own distribution and estimating the parameters using the (kernel) method and testing the goodness of fit of the model in general by proposing a test that relies primarily on the Fisher distribution. This was applied to medical data as a field of practical application, related to individuals who were exposed to a heart attack with two variables, one of which represents the response variable (Y), which expresses the number of times the patient was exposed to a heart attack, and the other variable (X), which expresses the patient's age as an independent variable. The results showed that the Poisson model represented the data correctly by passing the estimated model to the goodness of fit test, in addition to the results of the estimated parameters that appeared to be statistically significant, in light of the values ​​of the comparison criteria that resulted from the estimation results of the Poisson regression models for the application samples, which indicates the presence of a positive significant effect of the patient's age on the expected number of times he is exposed to a heart attack, which shows that the patient's age has a significant effect on the number of times he is exposed to a stroke for patients hospitalized in Ibn Al-Nafis Hospital in Baghdad Governorate for the year 2024.