Dear R-Help:
As you can see, Prof. Friendly refers me to your site for an executable version of vcd. I don't mean to be obtuse, but 15 minutes spent exploring your site failed to locate a downloadable version of the vcd package to which he referred.
I know plainly what this application can do. What I need to know is how to obtain the application itself.
My thanks in advance for any help you can provide.
MS
Well, vcd is a contributed package, not a stand-alone executable.
At first, you need to intstall R, after that, just type
install.packages("vcd")
if you are connected to the internet, and the package will be installed.
Then, you can use the function:
library(vcd)
goodfit(.....)But you will need to learn some basics of R at first, e.g. from "An Introduction to R" (a manual that comes with R), in order to import your data etc.
Uwe Ligges
------- Forwarded message follows ------- Date sent: Tue, 23 Sep 2003 09:14:28 -0400 From: Michael Friendly <[EMAIL PROTECTED]> Subject: Re: goodfit macro To: [EMAIL PROTECTED] Send reply to: [EMAIL PROTECTED] Organization: York University
You can also carry out goodness of fit tests using the function goodfit in the vcd package
for R (a free version of S/Splus),
http://www.r-project.org
-Michael
Michael Shott wrote:
Dear Michael:
Recently I searched the web for guidance in gauging the goodness-of-
fit of frequency data to statistical models. Your website came up in virtually every combination of keywords searched in Google. The site describes the goodfit.sas macro, which seems to do exactly what I'd like. But it seems to be available only as a .sas file. Do you have some application file that can be downloaded and used?
I ask because I'd like to analyze some archaeological data. They are frequency distributions of the number of a particular artifact type by site. That is, x sites may have 1 occurrence, y sites may have 2 occurrences, z sites 3 occurrences and so on. Your goodfit macro seems to do the job, but I can't execute it from the .sas version that appears on your website's ftp link.
Thanks for any help that you can provide.
Best,
Mike Shott
Michael J. Shott Professor Dept. of Sociology, Anthropology & Criminology University of Northern Iowa Cedar Falls, IA 50614-0513 319/273-7337
______________________________________________ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
