Variable selection in Logistic regression model using modified firefly algorithms

The logistic regression model is considered the most widely used in many applications, and it is one of the main models in the family of generalized linear models. Like other regression models, the model may contain many independent variables, which negatively affects the accuracy of the model and its simplicity in interpreting the results. This study aims to use the modified firefly algorithm and compare it with other methods for selecting variables in an exponential regression model using simulation and real data. The results showed that compared to other previously used methods, the proposed method performs better and helps reduce the mean square error of the model. .