[R list package] generating p values in ict.test

[Kosuke Imai]

Hi Shane,

  Try 

print(test.value.race)

instead of "summary()".  As for the second error, does your data set have a zero observation in one of the cells?  

Kosuke

Department of Politics
Princeton University
http://imai.princeton.edu

On Jan 3, 2013, at 6:19 PM, Shane Gleason <sgleason at siu.edu>
 wrote:

> Hello everyone,
> 
> We're using the list package for a project we're working on, but we're having some trouble with the ict.test command.
> 
> When we run through the examples provided, we cannot find anyway to display the standard error:  Specifically, we run:
> 
> data(race)
> 
> test.value.race <- ict.test(race$y, race$treat, J = 3, gms = TRUE)
> 
> This does not return anything- so we try:
> 
> summary(test.value.race)
> 
> This returns:
> 
>         Length Class  Mode
> p         1     -none- numeric
> pi.table 16     -none- numeric
> 
> 
> 
> I can't seem to figure out how to display the p value. Should I read that as the p value=1 (i.e. no design effect?)
> 
> 
> 
> On a somewhat related note, when I run the same test on our data I get the following error:
> 
> Error in solve.default(cov.pi.y1) :
>  system is computationally singular: reciprocal condition number = 0
> 
> 
> If I go on to call for summary(test.value) I get the same output as above.
> 
> Would this essentially be a non-starter? I tried to whittle down the data and drop variables we
> don't use and got the following error:
> 
> Error in t.test.default(y[treat == 0] <= j, y[treat == 1] <= j, alternative = "less") :
> 
>  data are essentially constant
> 
> I am sorry if these are elementary questions, but I wasn't able to find much guidance in the user manual.
> 
> Thanks for your time,
> 
> Shane
> 
> 
> 
> 
> 
> -- 
> Shane Gleason
> Doctoral Candidate
> Southern Illinois University:Carbondale
> Department of Political Science
> Faner 3172
> 
> _______________________________________________
> listpackage-discuss mailing list
> listpackage-discuss at lists.r-forge.r-project.org
> https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/listpackage-discuss