[R] question about sign of probit estimates

[peter dalgaard]

On Jul 20, 2010, at 2:39 AM, Nita Umashankar wrote:

> Hello,
> I am getting some results from my Probit estimation in R that are in the
> opposite direction of what I hypothesized.  In sas, the default is
> probability that y=0 (instead of 1) so one needs to type the word
> "descending" to get P(y=1).  Is the same true for R?  Is the default set to
> P(0)?

Why not just try it?

> y <- c(1,0,0)
> glm(y~1,binomial(probit))

Call:  glm(formula = y ~ 1, family = binomial(probit)) 

Coefficients:
(Intercept)  
    -0.4307  

Intercept is less than zero so it's probit(P(1))==qnorm(1/3).

(BTW, I don't think "what SAS does" is a well-defined concept. There are three procedures that will do probit regression - fortunately PROC CATMOD will not... - and they each have more than one way of specifying the response variable.) 


> Thank you in advance.
> 
> Nita Umashankar
> 
> 	[[alternative HTML version deleted]]
> 
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.

-- 
Peter Dalgaard
Center for Statistics, Copenhagen Business School
Solbjerg Plads 3, 2000 Frederiksberg, Denmark
Phone: (+45)38153501
Email: pd.mes at cbs.dk  Priv: PDalgd at gmail.com