Fitting Cox Proportional Model and Poisson Regression Model to Data of Patients with Stomach Cancer in Erbil-Kurdistan/Iraq

Abstract

Abstract— This study aims to fitting two models where allow the response variable to be the length of time (months) to data of patients with stomach cancer; cox proportional model and Poisson regression model, for modeling and identifying the affecting factors of stomach cancer patients. The
study was conducted between January 1, 2016 until December 31, 2020 for all patients with stomach cancer at Nanakali Main Hospital for Cancer in the Kurdistan Region of Iraq - Erbil. 

The results indicated that, the models have not reached to the same variables that have an impact on our data of patients with stomach cancer data in Erbil city. Moreover, according to the results the Poisson regression fitted data set very well depending on the Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC) values, the best model will be identified by the ones with smaller values. The data analyses of stomach cancer are done by using statistical programs (Mat-lab V.14 , SPSS V. 25 and STATGRAPHICS V. 19).