Tell me more ×
Cross Validated is a question and answer site for statisticians, data analysts, data miners and data visualization experts. It's 100% free, no registration required.
up vote 0 down vote favorite

I'm looking for a package to do truncated regression in R. I found the truncreg and truncSP packages, but they seem to allow either a lower bound or an upper bound, but not both at the same time.

All my response values are bound between 0 and 1, so I need to have both bounds taken into account.

Any help? Thanks a lot!

share|improve this question
1  
What about a beta regression? – Dason May 21 '12 at 0:33
Thanks for the suggestion. But is beta regression appropriate when the values are not ratios? (the underlying distribution is not beta) – Julián Urbano May 21 '12 at 0:56
Beta doesn't come from ratios necessarily. Are you thinking of binomial? I mean of course you would want the conditional distribution of the response given the predictors to actually be approximately beta to use beta regression but the beta distribution is quite flexible. – Dason May 21 '12 at 1:01
@Dason You're right. What I meant is what if the distribution is clearly not beta? It looks nearly uniform actually. Thanks again. – Julián Urbano May 21 '12 at 10:51
Can you describe what you mean by that more? Uniform on (0,1)? That's a special case of the beta. But if that's the case I don't understand why you'd be doing regression if it was uniform over the entire support of the predictors. – Dason May 21 '12 at 14:22
show 2 more comments

migrated from stackoverflow.com May 22 '12 at 15:22

Know someone who can answer? Share a link to this question via email, Google+, Twitter, or Facebook.

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.