Assessing logistic and poisson regression model for analyzing data count of patients with tuberculosis disease in Erbil, Iraqi Kurdistan region

Abstract EN

Many studies choose to analyze and classify count variables as binary.

The main purpose of this research is to determine how several predictor variables and a response variable relationship to each other’s.

To achieve this, logistic and Poisson regression models were being used.

The review of patient data count for TB disease in Erbil, Iraqi Kurdistan Region, is the main focus of this research.

where Gender (male, female) is a categorical response variable and there are multiple predictor variables.

The data collected on this disease indicates that it might be a big problem in our society, as it affects a wide range of people for a number of reasons.

The total number of cases during that time was (1346).

(2012 to 2018).

To assess each model's goodness of fit, two criterions were used.

The results showed that the Logistic regression model is the best fit in modeling binary response variable in the form of a count data based on the two evaluation criteria used [Akaike Information Criterions (AIC) and Bayesian Information Criterions (BIC)].

Microsoft Excel, Stata 16, and SPSS 25 are the statistical application software used.