[R] Log likelihood question

From: Phil (P.Atkinson@cdsc.nthames.nhs.uk)
Date: Fri 20 Jul 2001 - 02:11:44 EST

• Next message: Sven Garbade: "[R] Correction of degrees of freedom in repeated measure aov"
• Previous message: Zsombor Cseres-Gergely: "Re: [R] after R 1.3: can not write to or create directory `~/lib/ R'"

Message-id: <24A8B549739BD311AAF2009027D6373D198B84@SERVER>

I'm comparing a possion regression model (using glm) under R with same model
(using poisson) and same data in Stata.

Poisson GLM iteration using R has what it calls "deviance" values
Deviance = 1711.706 Iterations - 1
Deviance = 1451.352 Iterations - 2
Deviance = 1442.306 Iterations - 3
Deviance = 1442.283 Iterations - 4

Poisson GLM iteration using Stata gives log likelihood values
(0) -1567
(1) -1566
(2) -1566

As you can see, the R deviance values arent -2 * log likelihood values in
Stata. Any idea how R calculates its deviance values?

Many thanks..

Phil

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !) To: r-help-request@stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

• Next message: Sven Garbade: "[R] Correction of degrees of freedom in repeated measure aov"
• Previous message: Zsombor Cseres-Gergely: "Re: [R] after R 1.3: can not write to or create directory `~/lib/ R'"

This archive was generated by hypermail 2.1.3 : Wed 10 Mar 2004 - 08:18:20 EST