hurdle.IV: Hurdle IV regression

In jackiemauro/hurdleIV:

Description

Run a hurdle IV regression, either lognormal IV or cragg IV. You should specify the model you want to fit. Currently this does not have a nice summary function for the output.

Usage

hurdle.IV(formula, inst, endog, exog, data, endog_reg = list(),
  start_val = list(), type = "lognormal", options = list(cholesky = T,
  maxit = 5000, trace = 0, method = "BFGS"))

Arguments

formula	the second stage regression: y~exogenous + endogenous
inst	a vector of your instrument(s): c(inst1,inst2)
endog	a vector of your endogenous variable(s): c(end1,end2)
exog	a vector of your exogenous variable(s): c(ex1,ex2)
data	the dataframe
endog_reg	a list of endogenous regression formulae. By default will estimate endogenous ~ all exogenous variables + all instruments
start_val	an optional list of start values. By default the function will find start values using simple linear/probit regressions of the specified formulae
type	either "lognormal" or "cragg"
options:	cholesky true or false; maxit = maximal number of iterations; trace = 0; method = "BFGS". Options are similar to those for optim, see optim documentation for more details

Value

Returns estimated parameters as well as a the hessian and standard deviations

jackiemauro/hurdleIV documentation built on May 18, 2019, 7:56 a.m.