The negative binomial-Lindley generalized linear model: Characteristics and application using crash data

► We compared the negative binomial-Lindley model (NB-L) with the negative binomial (NB) and zero-inflated models. ► We used a simulated dataset and two observed datasets to evaluate the performance of the models. ► For the simulated dataset, the NB-L model was able to reproduce the “true” parameter values. ► The NB-L model offers the advantage of being able to handle datasets with a large number of zeros and/or high dispersion. ► The NB-L model provides a superior performance over the NB and zero-inflated models.