Introduction to the vol. 46, no. 2, 2019

The rationale of the Bayesian model checking method is to compare the discrepancy measure that calculates with the observed data to a distribution obtained by applying it to multiple simulated data sets. (2019) addresses one of the state-of-the-art methods for event-count time-series data analyses, the Poisson exponentially weighted moving average (P-EWMA) model. Simulations show that when the data are generated by a P-EWMA model, but an non-confounding covariate is omitted at the stage of estimation, the P-EWMA model’s credible interval is optimistically too narrow to contain the true value at the nominal level, whereas the NB-I(1) model does not suffer this problem.