Skip to main content

View full-text article in PMC

. 2018 Jan 5;13(1):e0190882. doi: 10.1371/journal.pone.0190882

Table 2. Goodness of fit evaluation to select the best model for count data.

Dependent variable	α value	LR test	Statistic	ZIP	ZINB
Total potential DDIs	0.38	LR test: chibar2 = 204.28 (p<0.001)	BIC	3943.813	3526.375
Total potential DDIs	0.38	LR test: chibar2 = 204.28 (p<0.001)	AIC	3828.689	3406.454
Serious potential DDIs	1.61 X 10⁻⁸	LR test: chibar2 = 1.8 X 10⁻⁵ (p = 0.4983)	BIC	482.589	488.357
Serious potential DDIs	1.61 X 10⁻⁸	LR test: chibar2 = 1.8 X 10⁻⁵ (p = 0.4983)	AIC	399.686	401.686
Significant potential DDIs	0.37	LR test: chibar2 = 131.57 (p<0.001)	BIC	1819.443	1694.099
Significant potential DDIs	0.37	LR test: chibar2 = 131.57 (p<0.001)	AIC	1726.416	1596.843
Minor potential DDIs	0.039	LR test: chibar2 = 1.15 (p = 0.1414)	BIC	1142.09	1147.059
Minor potential DDIs	0.039	LR test: chibar2 = 1.15 (p = 0.1414)	AIC	1051.395	1052.242

ZIP: Zero-inflated Poisson model.

ZINB: Zero-inflated negative binomial model.

BIC: Bayesian information criteria.

AIC: Akaike information criteria.

LR test: Likelihood ratio test.