[R-SIG-Mac] Problem running TOBIT for R 3.2 at SPSS 24 Interface MAC Sierra 10.12.1

[Peter Dalgaard]

It really should be SPSS's responsibility to track these issues. They are sponging off our work, precariously close to overstepping the license conditions, creating new problems, then expecting us to come and fix them.

As for the concrete issue, you seem to be bitten by what was a major annoyance in the 3.2 series: pbkrtest requiring R >= 3.2.3 and hence all packages depending on it (like car) doing so as well. 

If SPSS hardcodes the R version, you are up the proverbial creek... You might get away with hacking the DESCRIPTION file for pbkrtest to make it install (not work, but you likely won't use it anyway) under 3.2.0. Of course, if it breaks you can keep both pieces.

I'd probably be more inclined to save out the data set, fire up an appropriate version of R (or RStudio), do

library(foreign)
d <- read.spss("myfile.sav", to.data.frame=TRUE)
library(AER)
....

and so forth.

You may need to study the R syntax a little. Achim and Christian wrote an inexpensive book called "Applied Econometrics with R"....

-pd

> On 8 Dec 2016, at 10:34 , Andreas Schröder <andreas.j.schroeder at gmail.com> wrote:
> 
> Dear R-SIG-Mac members, 
> 
> This is my first post and I would like to thank you first of all for helping out fellow users of R!
> 
> I am currently doing my master thesis and struggle with making the TOBIT regression run on my SPSS 24, operated with Mac (OS Sierra 10.12.1). 
> 
> I use SPSS 24 and have the installer plugin-in for R successfully  installed. I installed the correct R version 3.2 and also XQuartz but if I try to execute the TOBIT regression (I can choose it in my SPSS menu now) I get constantly dialogues showing that certain packages would be missing (CAR, AED etc.). (Pic 1). Apparently, those packages have been added later to R (higher r version than the one mandated by my spss 24 version). I downloaded the versions R 3.2.5 that is referred to in the first picture but once I try installing it, I get displayed another error message by SPSS that the version of R was incorrect. Also manually installing the AER package on R 3.2. did not work for me because I do not know where to find and executable file within the AER package. (Pic 3)
> 
> 
> I already spent yesterday and partly the day before to find my mistake but I couldn’t resolve the issue.
> I am glad to receive any kind of input!
> 
> Best, 
> 
> Andreas
> 
> <Dialogue after Tobit.png><Error Message.png><AER installation.png>
> _______________________________________________
> R-SIG-Mac mailing list
> R-SIG-Mac at r-project.org
> https://stat.ethz.ch/mailman/listinfo/r-sig-mac