The new Log-Linear Test (TL) is proposed to identify when the Poisson model fails for a collection of count random variables. TL is shown to have better rejection rate with small sample size and essentially the same power compared to a classical Fisher-Bohning’s Statistic TF for standard alternatives to Poisson.
Chen, Haiyan and Stratton, Howard H.
"Test for Spatio-Temporal Counts Being Poisson,"
Journal of Modern Applied Statistical Methods:
1, Article 22.
Available at: http://digitalcommons.wayne.edu/jmasm/vol7/iss1/22