Estimation of nested and zero-inflated ordered probit models

We introduce three new commands—nop, ziop2, and ziop3—for the estimation of a three-part nested ordered probit model, two-part zero-inflated models Harris Zhao (2007, Journal Econometrics 141: 1073–1099) Brooks, Harris, Spencer (2012, Economics Letters 117: 683–686), model Sirchenko (2020, Studies in Nonlinear Dynamics 24: 1) for ordinal outcomes, with both exogenous endogenous switching. The allow probabilities positive, neutral (zero), negative outcomes to be generated by distinct processes. address preponderance zeros them emerge different latent regimes. provide postestimation commands compute probabilistic predictions various measures their accuracy, assess goodness fit, perform comparison using Vuong test (Vuong, 1989, Econometrica 57: 307–333) corrections based on Akaike Schwarz information criteria. investigate finite-sample performance maximum likelihood estimators Monte Carlo simulations, discuss relations among models, illustrate an empirical application U.S. federal funds rate target.