Take the 2-minute tour ×
Cross Validated is a question and answer site for people interested in statistics, machine learning, data analysis, data mining, and data visualization. It's 100% free, no registration required.

I use the locfit package for some local logistic regression in R, and its scb function to plot simultaneous confidence bands. However, I see no way to request 90% confidence bands instead of the standard 95% ones. I checked the functions themselves, but I do not see where one could hack those either.

Why is there this limitation? What can one do about this?

My code goes along these lines:

fit2<-scb(closed_rule ~ bl,deg=1,type=4,
 xlim=c(0,1),ev=lfgrid(100), 
 family='binomial',data=data,alpha=cbind(0,0.3))
share|improve this question
1  
Please DO NOT cross-post. –  mbq Oct 10 '11 at 22:52
    
Understood. Which place do you think is better for similar, R-related questions when it is not really about a statistical method, but "programming"? –  László Oct 10 '11 at 23:32
    
I think this question is perfectly fine here. If you use stackexchange, you will see why cross-posting is futile. Have a look at my setting: stackexchange.com/filters/19452/tuto-completo –  Roman Luštrik Oct 11 '11 at 9:26
    
@László Opinions vary, but in general the ideal location seems to vary based on the degree to which it is purely about R (Stackoverflow) versus being about statistics (Cross validated). There's a set of questions that fill the middle ground where either site is fine. In general, post once, and see what response you get; if you don't get a good response after a couple of days and decide to cross post, always provide a link your questions with a url. –  Jeromy Anglim Oct 11 '11 at 11:35

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.