Hi All, I'm using glm() in R to perform Poisson regression, I'm wondering if its possible to get equivalent Type 1 / Type 3 Analysis (similar to one in PROC GENMOD). Thanks, Kim [[alternative HTML version deleted]]
Bill.Venables at csiro.au
2010-Jan-28 03:57 UTC
[R] SAS Type 1 / Type 3 Analysis Equivalent.
Why do you think we speak SAS here? anova(fm, test = "Chisq") will get you a sequential analysis of variance table. library(MASS) dropterm(fm, test = "Chisq") will show you the effect of removing the *non-marginal* terms one at a time. What more does anyone need? Bill Venables CSIRO/CMIS Cleveland Laboratories -----Original Message----- From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-project.org] On Behalf Of Kim Jung Hwa Sent: Thursday, 28 January 2010 1:44 PM To: r-help at r-project.org Subject: [R] SAS Type 1 / Type 3 Analysis Equivalent. Hi All, I'm using glm() in R to perform Poisson regression, I'm wondering if its possible to get equivalent Type 1 / Type 3 Analysis (similar to one in PROC GENMOD). Thanks, Kim [[alternative HTML version deleted]] ______________________________________________ R-help at r-project.org 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.
Dear Kim,
The standard R function anova() produces sequential ("type I") tests;
the
Anova() function in the car package can produce "type-III" tests (it
performs "type-II" tests by default), but you must be careful with
contrast
coding if there are terms in the model related by marginality.
I hope this helps,
John
--------------------------------
John Fox
Senator William McMaster
Professor of Social Statistics
Department of Sociology
McMaster University
Hamilton, Ontario, Canada
web: socserv.mcmaster.ca/jfox
> -----Original Message-----
> From: r-help-bounces at r-project.org [mailto:r-help-bounces at
r-project.org]
On> Behalf Of Kim Jung Hwa
> Sent: January-27-10 10:44 PM
> To: r-help at r-project.org
> Subject: [R] SAS Type 1 / Type 3 Analysis Equivalent.
>
> Hi All,
>
> I'm using glm() in R to perform Poisson regression, I'm wondering
if its
> possible to get equivalent Type 1 / Type 3 Analysis (similar to one in
PROC> GENMOD).
>
> Thanks,
> Kim
>
> [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at r-project.org 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.