Anticipated Date of Graduation

Summer 2022

Document Type

Thesis

Degree Name

Master of Science in Mathematical Sciences

Department

Mathematical Sciences

First Advisor

Douglas Darbro

Abstract

The type I error rates for the binary logistic regression model were examined across varying levels of multicollinearity. Population data sets were created using the statistical software package R and then used to create data suitable for binary logistic regression models. The results showed the type I error rates did not differ across multicollinearity levels, the percentage of accurately classified cases were unaffected, and the variance inflation factors were affected by more than just correlation between the independent variables. These results show that multicollinearity may have limited effects on the type I error rates of the binary logistic regression model; however, these effects should not be ignored.

Included in

Mathematics Commons

Share

COinS