[#307] block variances incorrectly scaled in quasi-poisson fits

Date:
2009-02-17 16:39 Priority:
3

State:
Closed

Submitted by:
Ben Bolker (bbolker) Assigned to:
Nobody (None)

Product:
glmer Operating System:
All

Component:
R or C functions Version:
None

Severity:
normal Resolution:
None

URL:

Summary:
block variances incorrectly scaled in quasi-poisson fits

Detailed description
I believe, based on some reasonably careful testing, that quasi- fits are incorrectly scaling the reported random effects variance by the error (residual) variance. A detailed discussion is given by: http://www.zoology.ufl.edu/bolker/glmm/quasitest.Rnw http://www.zoology.ufl.edu/bolker/glmm/quasitest.pdf -- essentially, I generate some simulated data and show that quasi- are inflating the block variances inappropriately. I believe the following patch fixes the problem: --- lmer.R (revision 279) +++ lmer.R (working copy) @@ -938,14 +938,19 @@ else "sigmaML"] }) +is.glmm <- function(object) { + length(object@muEta)>0 +} + setMethod("VarCorr", signature(x = "mer"), function(x, ...) ### Create the VarCorr object of variances and covariances { - sc <- sigma(x) + tsc <- sc <- sigma(x) + if (is.glmm(x)) tsc <- 1 ## DON'T scale GLMM var-corr by ans <- lapply(cc <- .Call(mer_ST_chol, x), function(ch) { - val <- crossprod(sc * ch) # variance-covariance + val <- crossprod(tsc * ch) # variance-covariance stddev <- sqrt(diag(val)) correl <- t(val / stddev)/stddev diag(correl) <- 1

Comments:

Message
Date: 2012-03-17 22:35 Sender: Ben Bolker quasi-poisson no longer exists, this is obsolete!

Field	Old Value	Date	By
status_id	Open	2012-03-17 22:35	bbolker
close_date	None	2012-03-17 22:35	bbolker
details	I believe, based on some reasonably careful testing, that quasi- fits are incorrectly scaling the reported random effects variance by the error (residual) variance. A detailed discussion is given by: http://www.zoology.ufl.edu/bolker/glmm/quasitest.Rnw http://www.zoology.ufl.edu/bolker/glmm/quasitest.pdf -- essentially, I generate some simulated data and show that quasi- are inflating the block variances inappropriately. I believe the following patch fixes the problem: --- lmer.R (revision 279) +++ lmer.R (working copy) @@ -938,14 +938,19 @@ else "sigmaML"] }) +is.glmm <- function(object) { + length(object@muEta)>0 +} + setMethod("VarCorr", signature(x = "mer"), function(x, ...) ### Create the VarCorr object of variances and covariances { - sc <- sigma(x) + tsc <- sc <- sigma(x) + if (is.glmm(x)) tsc <- 1 ## DON'T scale GLMM var-corr by ans <- lapply(cc <- .Call(mer_ST_chol, x), function(ch) { - val <- crossprod(sc * ch) # variance-covariance + val <- crossprod(tsc * ch) # variance-covariance stddev <- sqrt(diag(val)) correl <- t(val / stddev)/stddev diag(correl) <- 1	2012-03-17 22:35	bbolker