R help - Sep 2011 - linear probability model

Dear All,

Is there any function in R for fitting linear probability model, as my
response variable is a uniformly distributed.

Regards,
Adam

	[[alternative HTML version deleted]]

Hi,
You might want to take a look at ?glm with logit link function.
JC

2011/9/14 Adam Copella <adam.copella at
gmail.com>:
> Dear All,
>
> Is there any function in R for fitting linear probability model, as my
> response variable is a uniformly distributed.
>
> Regards,
> Adam
>
> ? ? ? ?[[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.
>

On Wed, 14 Sep 2011, Adam Copella wrote:
> Dear All,
>
> Is there any function in R for fitting linear probability model, as my
> response variable is a uniformly distributed.
If it is really uniformly distributed, I think there is not much to 
model... And in the textbooks I know that discuss the linear probability 
model (mostly econometrics books), this typically refers to running OLS on 
a 0/1 response variable. This can be done with lm() but is typically 
inappropriate and using a binomial glm() is a better solution (as 
previously pointed out).

However, from your description it seems that your response takes values in 
the open unit interval and that you want to model its distributional 
properties. Then beta regression may be an option. See package
"betareg"
and the accompanying paper http://www.jstatsoft.org/v34/i02/.

hth,
Z
> Regards,
> Adam
>
> 	[[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.
>