A Comparative Study of Selected Regression Models Using Road Traffic Crashes Data in Ekiti State
- September 18, 2020
- Posted by: RSIS Team
- Categories: IJRIAS, Mathematics
International Journal of Research and Innovation in Applied Science (IJRIAS) | Volume V, Issue III, January 2020 | ISSN 2454–6186
A Comparative Study of Selected Regression Models Using Road Traffic Crashes Data in Ekiti State
Odukoya Elijah Ayooluwa, Olubiyi Adenike Oluwafumilola
Department of Statistics, Ekiti State University, Nigeria
Abstract – Road traffic crashes are count (discrete) in nature, when modelling discrete data for characteristics and prediction of an event when dependent variable are non-negative and have integer values, it is appropriate to use Poisson regression. However the condition that the mean and variance of Poisson are equal poses a great constraints. Data on road traffic crashes from Federal Road Safety Commission (FRSC) Ekiti state Nigeria were analyzed using R software package. The result from the three existing model were compared using AIC, BIC and Deviance, with Generalized negative binomial showing an AIC value of 414.79 and BIC value of 490.8873 and negative binomial showing AIC value of 476.8 and BIC value of 495.59 and Poisson regression showing AIC value of 587.312 and BIC value of 589.321.Having shown a smaller value of AIC and BIC, Generalized negative binomial regression was consider a better model when analyzing road traffic crashes in Ekiti State Nigeria.
Keywords-Poisson regression model, Negative binomial regression, Generalized negative binomial regression.
