summary.manyglm fails if family is "binomial(link=logit)"

[raymondben]

Fails:
[from the demo code]

glm.spid.bin <- manyglm(pres.abs~soil.dry+bare.sand+moss, data=X, family="binomial")
summary(glm.spid.bin)

Error in summary.manyglm(glm.spid.bin) :
'family' not defined. Choose one of 'poisson', 'negative.binomial', 'binomial' for an manyglm object

The family is:

glm.spid.bin$family
[1] "binomial(link=logit)"

But summary.manyglm looks for an exact string match on the family (object$family == "binomial"), so it fails. Probably this test should be more like the one used in anova.manyglm (substr(object$family, 1, 1) == "b")

sessionInfo()
R version 3.2.2 (2015-08-14)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows 7 x64 (build 7601) Service Pack 1
attached base packages:
[1] stats graphics grDevices utils datasets methods base
other attached packages:
[1] mvabund_3.11.5

[aliceyiwang]

Hi, Ben,

I don't get the error in summary.manyglm(glm.spid.bin) by running the
example. I'm not sure which version you are looking at, but the line
(substr(object$family,
1, 1) == "b") is not in anova.manyglm() on GitHub. In fact, the substring
function is used only in manyglm() to parse the input argument 'family' to
object$family. Both summary and anova functions inherit the object's family
class from object$family. Therefore, unless the attribute is manually
altered, you should not get the error with summary.manyglm or
anova.manyglm. Please let me know if there are still problems.

Alice

On Fri, Mar 18, 2016 at 10:33 AM, Ben Raymond notifications@github.com
wrote:

Fails:
[from the demo code]
glm.spid.bin <- manyglm(pres.abs~soil.dry+bare.sand+moss, data=X,
family="binomial")
summary(glm.spid.bin)

Error in summary.manyglm(glm.spid.bin) :
'family' not defined. Choose one of 'poisson', 'negative.binomial',
'binomial' for an manyglm object

The family is:

glm.spid.bin$family
[1] "binomial(link=logit)"

But summary.manyglm looks for an exact string match on the family
(object$family == "binomial"), so it fails. Probably this test should be
more like the one used in anova.manyglm (substr(object$family, 1, 1) == "b")

sessionInfo()
R version 3.2.2 (2015-08-14)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows 7 x64 (build 7601) Service Pack 1

attached base packages:
[1] stats graphics grDevices utils datasets methods base

other attached packages:
[1] mvabund_3.11.5

—
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub
#20

[raymondben]

We were using the CRAN version (3.11.5, see my sessionInfo output above). You're right, the current dev version (3.11.7) doesn't have this problem. Sorry, didn't check that before submitting. Thanks.

[eddelbuettel]

Thanks for the follow-up. Maybe time for Alice to roll up a new version...

[aliceyiwang]

David,

Maybe it is time to update the CRAN version?

Thanks,
Alice

On Mon, Mar 21, 2016 at 5:59 AM, Dirk Eddelbuettel <notifications@github.com

wrote:

Thanks for the follow-up. Maybe time for Alice to roll up a new version...

—
You are receiving this because you commented.
Reply to this email directly or view it on GitHub
#20 (comment)