dear all, for some reasons I would like to add some new components (vectors, matrices) to a "glmerMod" fit. I am not familiar with S4, so borrowing from S3 classes my first tentative was o <- glmer(y ~ ..) o at new<-1:3 But it clearly does not work. Is there any solution? thanks in advance for your time, best, vito
adding objects into "glmerMod" fits
3 messages · vito muggeo, Ben Bolker, Henrik Singmann
Unfortunately for you, S4 classes are much fussier than S3 classes
about their composition. Your only choices, I think, are
(1) modify the definition of a glmerMod (by copying the entire code of
the package, or forking it on Github)
(2) construct a class that extends the glmerMod (i.e with
setClass("myGlmerMod", contains="glmerMod"))
e.g.
setClass("myGlmerMod",contains="glmerMod",representation(new="data.frame"))
seems to work - at least it doesn't complain.
(3) add the additional information as one or more attributes
On Tue, Oct 3, 2017 at 5:39 PM, Vito Michele Rosario Muggeo
<vito.muggeo at unipa.it> wrote:
dear all, for some reasons I would like to add some new components (vectors, matrices) to a "glmerMod" fit. I am not familiar with S4, so borrowing from S3 classes my first tentative was o <- glmer(y ~ ..) o at new<-1:3 But it clearly does not work. Is there any solution? thanks in advance for your time, best, vito
_______________________________________________ R-sig-mixed-models at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
FWIW, the approach of extending the class via setClass is also the one taken by lmerTest. And lmerTest works quite well. See: https://github.com/cran/lmerTest/blob/master/R/classes.R Best, Henrik Am 04.10.2017 um 00:41 schrieb Ben Bolker:
Unfortunately for you, S4 classes are much fussier than S3 classes
about their composition. Your only choices, I think, are
(1) modify the definition of a glmerMod (by copying the entire code of
the package, or forking it on Github)
(2) construct a class that extends the glmerMod (i.e with
setClass("myGlmerMod", contains="glmerMod"))
e.g.
setClass("myGlmerMod",contains="glmerMod",representation(new="data.frame"))
seems to work - at least it doesn't complain.
(3) add the additional information as one or more attributes
On Tue, Oct 3, 2017 at 5:39 PM, Vito Michele Rosario Muggeo
<vito.muggeo at unipa.it> wrote:
dear all, for some reasons I would like to add some new components (vectors, matrices) to a "glmerMod" fit. I am not familiar with S4, so borrowing from S3 classes my first tentative was o <- glmer(y ~ ..) o at new<-1:3 But it clearly does not work. Is there any solution? thanks in advance for your time, best, vito
_______________________________________________ R-sig-mixed-models at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models