[R] Simultaneous Probit-Tobit

[David Winsemius]

On Feb 22, 2015, at 3:17 PM, Christopher Shultz wrote:

> Hi: 
> 
> 
> 
> 
> I am trying to write code to run a simultaneous probit-tobit 
> model (Chappell, 1982) in order to quantify the determinants of 
> legislator voting behavior. I can't seem to find any background on this.
> I have worked with code for simple logit and probit models, but this is
> a different beast. 
> 
> The logic is that this model reduces some of the endogeneity 
> associated with a single equation model. As a general overview of the 
> structure: http://uploadpie.com/yB5mw

I'm pretty much clueless about "endogeneity", but I do know that Therneau uses this as an example in survival::survreg:

# Economists fit a model called `tobit regression', which is a standard
# linear regression with Gaussian errors, and left censored data.
tobinfit <- survreg(Surv(durable, durable>0, type='left') ~ age + quant,
	            data=tobin, dist='gaussian')

> Does anyone know if there is a package or simple way to code this? Thanks in advance! 

When I ask (or am asked) the question: is there code to do X? .... I reach for the sos-package:

install.packages('sos')
library(sos)
findFn("tobit")  # 98 links
findFn("tobit probit")
found 43 matches;  retrieving 3 pages
2 3 
Downloaded 25 links in 11 packages.

>    		 	   		  
> 	[[alternative HTML version deleted]]
> 
Please read the posting guide.

-- 

David Winsemius
Alameda, CA, USA