Stata 15 help for j_vuong

Vuong test is not appropriate for testing zero inflation

Greene (1994) proposed using the Vuong test for nonnested models (Vuong 1989) to test for zero inflation. Despite many earlier citations, recent work by Wilson (2015) has shown that the Vuong test is inappropriate for testing zero inflation. Nesting occurs when the probability of zero inflation is 0, which is on the boundary, and this violates the regularity conditions of the Vuong test for nonnested models. So the distribution of the test statistic is not standard normal. The actual distribution is unknown and thus cannot be used for inference.

You may consider using information criteria to choose between the standard and the zero-inflated models; see example 2 in [R] zip.

References

Greene, W. H. 1994. Accounting for excess zeros and sample selection in Poisson and negative binomial regression models. Working paper EC-94-10, Department of Economics, Stern School of Business, New York University. https://ideas.repec.org/p/ste/nystbu/94-10.html.

Vuong, Q. H. 1989. Likelihood ratio tests for model selection and non-nested hypotheses. Econometrica 57: 307-333.

Wilson, P. 2015. The misuse of the Vuong test for non-nested models to test for zero-inflation. Economics Letters 127: 51-53.


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index