Skip to main content
. 2020 Jul 1;20:175. doi: 10.1186/s12874-020-01055-2

Fig. 9.

Fig. 9

Comparison of the type I errors and powers of the SW tests for the RQR, MQR, deviance residual and Pearson residual. Response variable is simulated from the true model at varying sample size, n=20,50,100,200,400,600,800,1000 and percentage of excessive zeroes of p=10% (Inline graphic), 30% (Inline graphic) and 50% (Inline graphic). True model: ZIP model with mean exp(β0+β1x). Wrong model: Poisson model with mean exp(β0+β1x)