Multiple Hurdle Tobit Models

Estimation of models with dependent variable left-censored at zero. Null values may be caused by a selection process Cragg (1971) , insufficient resources Tobin (1958) , or infrequency of purchase Deaton and Irish (1984) .


Overview

Data collected by means of households' expenditure survey may present a large proportion of zero expenditures due to many households recording, for one reason or another, no expenditure for some items. Since the seminal paper of Tobin (1958), a large econometric literature has been developed to deal correctly with this problem of zero observations. In particular, a good selection mechanism was introduced by Cragg (1971) and a purchasing mechanism by Deaton and Irish (1984). We propose an encompassing approach with a general three equations model for which a zero expense can be observed either because:

  • the good is not selected,
  • the good is selected but not consumed because of lack of financial resources,
  • the good is selected and consumed, but purchased infrequently so that no expense is observed during the period of the survey.

mhurdle provides a set of tools to estimate (by maximum likelihood) and test (using especially vuong test) this generalized hurdle model.

Installation

mhurdle is on CRAN.

install.packages("mhurdle")

For the development version, use

install.packages("devtools")
devtools::install_github("ycroissant/mhurdle")

Reference manual

It appears you don't have a PDF plugin for this browser. You can click here to download the reference manual.

install.packages("mhurdle")

1.3-2 by Yves Croissant, a year ago


https://www.R-project.org


Browse source code at https://github.com/cran/mhurdle


Authors: Yves Croissant [aut, cre] (ORCID: , Fabrizio Carlevaro [aut] , Stephane Hoareau [aut]


Documentation:   PDF Manual  


GPL (>= 2) license


Imports Formula, truncreg, maxLik, survival, Rdpack, prediction, margins, generics, numDeriv, sandwich, nonnest2, CompQuadForm

Suggests knitr, rmarkdown, lmtest, bookdown, testthat, modelsummary, tibble, broom


Suggested by insight, marginaleffects.

Enhanced by texreg.


See at CRAN