[R] Predicted values from a logistic model

[Frank E Harrell Jr]

MANASI VYDYANATH wrote:
> Hello -
> 
> I apologize if this question is simple/obvious, but I couldn't find a  
> satisfactory answer online, and I am not very accustomed to working  
> with R (Matlab is my poison. :-)). Any help would be greatly  
> appreciated.
> 
> I have a model with a three-level factor and a continuous covariate.  
> The call I use is:
> 
> mymodel <- glm(Response ~ Factor_covariate + continuous_covariate -  
> 1, family = binomial(link = "logit"))
> 
> I would like to generate predicted values for a given level of the  
> covariate, and a given level of the factor. For instance, I want it  
> to give me a fitted value for the response at factor level 1 and  
> continuous covariate value 10. How would I go about expressing this?  
> I tried to look at the package Design, and specifically, at the  
> command "predict.lrt". But I was unable to quite understand how I  
> ought to enter my x-values.  Again, any help would be much appreciated.
> 
> Thank you for taking the time to read this!
> 
> Cheers,
> 
> Manasi

With Design you do predict(mymodel, data.frame(age=30, sex='male'), 
type='fitted')

For ordinal responses there are several options for prediction different 
things.  If you want to leave some covariates unspecified (default to 
reference values - medians or modes) you can use predict(mymodel, 
gendata(mymodel, list of covariates you care to specify))

Frank Harrell

> 
> ______________________________________________
> R-help at stat.math.ethz.ch 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.
> 


-- 
Frank E Harrell Jr   Professor and Chair           School of Medicine
                      Department of Biostatistics   Vanderbilt University