No support for meqrlogit as of Stata 16

[gunnu singh]

I looked at the help file for meqrlogit in Stata 16 and it is no longer part of Stata 16. I couldn't find any alternative to it in melogit and would like to know the alternative.

meqrlogit continues to work but, as of Stata 16, is no longer an official part of Stata.

See melogit for a recommended alternative to meqrlogit.

[Scott Merryman]

-melogit- has the option of using the Pinheiro-Chao mode-curvature adaptive Gauss-Hermite quadrature, which I believe is method used by -meqrlogit-.

[gunnu singh]

I used integration method mode-curvature adaptive Gauss-Hermite quadrature (mcaghermite) but it's giving error whereas meqrlogit ran fine. Is there anything I am missing?

[Joseph Coveney]

I used integration method mode-curvature adaptive Gauss-Hermite quadrature (mcaghermite) but it's giving error whereas meqrlogit ran fine. Is there anything I am missing?

Yeah: Scott's post. Please re-read it.

And then look at the help file for -melogit- where you'll find that the Pinheiro-Chao mode-curvature adaptive Gauss-Hermite quadrature method is not -mcaghermite-.

[gunnu singh]

Ok, I see Pinheiro-Chao mode-curvature adaptive Gauss-Hermite quadrature method in melogit but when I open help file from Stata (h melogit), this method is not defined there. However, I tried using this method and it gave error

method pcaghermite not recognized

.

[Joseph Coveney]

. . . when I open help file from Stata (h melogit), this method is not defined there. However, I tried using this method and it gave error .

That's funny; I copied and pasted the text in my post above directly from the help file for melogit. And the quadrature method works for me (see example output below—begin at the "Begin here" comment).

You might want to check if your Stata is up-to-date. I show how to test for that, too, in the output below—see output following the comment related to checking for update status. Or you can just check by inspection after executing the command.

.ÿ
.ÿversionÿ16.1

.ÿ
.ÿclearÿ*

.ÿ
.ÿsetÿseedÿ`=strreverse("1574672")'

.ÿquietlyÿsetÿobsÿ200

.ÿ
.ÿgenerateÿintÿpidÿ=ÿ_n

.ÿgenerateÿdoubleÿpid_uÿ=ÿrnormal()

.ÿ
.ÿquietlyÿexpandÿ2

.ÿbysortÿpid:ÿgenerateÿbyteÿtimÿ=ÿ_n

.ÿ
.ÿgenerateÿdoubleÿxbÿ=ÿpid_uÿ+ÿ0ÿ*ÿ(timÿ-ÿ1)

.ÿgenerateÿbyteÿoutÿ=ÿrbinomial(1,ÿinvlogit(xb))

.ÿ
.ÿ*
.ÿ*ÿBeginÿhere
.ÿ*
.ÿmelogitÿoutÿi.timÿ||ÿpid:ÿ,ÿintmethod(pcaghermite)ÿnolog

Mixed-effectsÿlogisticÿregressionÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿNumberÿofÿobsÿÿÿÿÿ=ÿÿÿÿÿÿÿÿ400
Groupÿvariable:ÿÿÿÿÿÿÿÿÿÿÿÿÿpidÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿNumberÿofÿgroupsÿÿ=ÿÿÿÿÿÿÿÿ200

ÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿObsÿperÿgroup:
ÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿminÿ=ÿÿÿÿÿÿÿÿÿÿ2
ÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿavgÿ=ÿÿÿÿÿÿÿÿ2.0
ÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿmaxÿ=ÿÿÿÿÿÿÿÿÿÿ2

Integrationÿmethod:ÿpcaghermiteÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿIntegrationÿpts.ÿÿ=ÿÿÿÿÿÿÿÿÿÿ7

ÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿWaldÿchi2(1)ÿÿÿÿÿÿ=ÿÿÿÿÿÿÿ1.16
Logÿlikelihoodÿ=ÿ-273.58317ÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿProbÿ>ÿchi2ÿÿÿÿÿÿÿ=ÿÿÿÿÿ0.2820
------------------------------------------------------------------------------
ÿÿÿÿÿÿÿÿÿoutÿ|ÿÿÿÿÿÿCoef.ÿÿÿStd.ÿErr.ÿÿÿÿÿÿzÿÿÿÿP>|z|ÿÿÿÿÿ[95%ÿConf.ÿInterval]
-------------+----------------------------------------------------------------
ÿÿÿÿÿÿÿ2.timÿ|ÿÿÿ.2336172ÿÿÿ.2171401ÿÿÿÿÿ1.08ÿÿÿ0.282ÿÿÿÿ-.1919696ÿÿÿÿÿ.659204
ÿÿÿÿÿÿÿ_consÿ|ÿÿ-.3032607ÿÿÿ.1670008ÿÿÿÿ-1.82ÿÿÿ0.069ÿÿÿÿ-.6305762ÿÿÿÿ.0240549
-------------+----------------------------------------------------------------
pidÿÿÿÿÿÿÿÿÿÿ|
ÿÿÿvar(_cons)|ÿÿÿ.7369954ÿÿÿ.4889082ÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿ.2008143ÿÿÿÿ2.704799
------------------------------------------------------------------------------
LRÿtestÿvs.ÿlogisticÿmodel:ÿchibar2(01)ÿ=ÿ3.78ÿÿÿÿÿÿÿÿProbÿ>=ÿchibar2ÿ=ÿ0.0259

.ÿ
.ÿ//ÿCheckÿyourÿupdateÿstatus
.ÿquietlyÿqueryÿupdate

.ÿforeachÿcomponentÿinÿexeÿadoÿutilitiesÿdocsÿ{
ÿÿ2.ÿÿÿÿÿassertÿr(inst_`component')ÿ==ÿr(avbl_`component')
ÿÿ3.ÿ}

.ÿ
.ÿexit

endÿofÿdo-file


.

[Nick Cox]

Seemingly, cross-posted at https://stackoverflow.com/questions/...as-of-stata-16

Please note our policy about cross-posting which is that you should tell us about it.

[gunnu singh]

Seemingly, cross-posted at https://stackoverflow.com/questions/...as-of-stata-16

Please note our policy about cross-posting which is that you should tell us about it.

I apologize that I missed this part of the policy. I cross-posted to get a wider audience.

[gunnu singh]

That's funny; I copied and pasted the text in my post above directly from the help file for melogit. And the quadrature method works for me (see example output below—begin at the "Begin here" comment).

You might want to check if your Stata is up-to-date. I show how to test for that, too, in the output below—see output following the comment related to checking for update status. Or you can just check by inspection after executing the command.

.ÿ
.ÿversionÿ16.1

.ÿ
.ÿclearÿ*

.ÿ
.ÿsetÿseedÿ`=strreverse("1574672")'

.ÿquietlyÿsetÿobsÿ200

.ÿ
.ÿgenerateÿintÿpidÿ=ÿ_n

.ÿgenerateÿdoubleÿpid_uÿ=ÿrnormal()

.ÿ
.ÿquietlyÿexpandÿ2

.ÿbysortÿpid:ÿgenerateÿbyteÿtimÿ=ÿ_n

.ÿ
.ÿgenerateÿdoubleÿxbÿ=ÿpid_uÿ+ÿ0ÿ*ÿ(timÿ-ÿ1)

.ÿgenerateÿbyteÿoutÿ=ÿrbinomial(1,ÿinvlogit(xb))

.ÿ
.ÿ*
.ÿ*ÿBeginÿhere
.ÿ*
.ÿmelogitÿoutÿi.timÿ||ÿpid:ÿ,ÿintmethod(pcaghermite)ÿnolog

Mixed-effectsÿlogisticÿregressionÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿNumberÿofÿobsÿÿÿÿÿ=ÿÿÿÿÿÿÿÿ400
Groupÿvariable:ÿÿÿÿÿÿÿÿÿÿÿÿÿpidÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿNumberÿofÿgroupsÿÿ=ÿÿÿÿÿÿÿÿ200

ÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿObsÿperÿgroup:
ÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿminÿ=ÿÿÿÿÿÿÿÿÿÿ2
ÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿavgÿ=ÿÿÿÿÿÿÿÿ2.0
ÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿmaxÿ=ÿÿÿÿÿÿÿÿÿÿ2

Integrationÿmethod:ÿpcaghermiteÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿIntegrationÿpts.ÿÿ=ÿÿÿÿÿÿÿÿÿÿ7

ÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿWaldÿchi2(1)ÿÿÿÿÿÿ=ÿÿÿÿÿÿÿ1.16
Logÿlikelihoodÿ=ÿ-273.58317ÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿProbÿ>ÿchi2ÿÿÿÿÿÿÿ=ÿÿÿÿÿ0.2820
------------------------------------------------------------------------------
ÿÿÿÿÿÿÿÿÿoutÿ|ÿÿÿÿÿÿCoef.ÿÿÿStd.ÿErr.ÿÿÿÿÿÿzÿÿÿÿP>|z|ÿÿÿÿÿ[95%ÿConf.ÿInterval]
-------------+----------------------------------------------------------------
ÿÿÿÿÿÿÿ2.timÿ|ÿÿÿ.2336172ÿÿÿ.2171401ÿÿÿÿÿ1.08ÿÿÿ0.282ÿÿÿÿ-.1919696ÿÿÿÿÿ.659204
ÿÿÿÿÿÿÿ_consÿ|ÿÿ-.3032607ÿÿÿ.1670008ÿÿÿÿ-1.82ÿÿÿ0.069ÿÿÿÿ-.6305762ÿÿÿÿ.0240549
-------------+----------------------------------------------------------------
pidÿÿÿÿÿÿÿÿÿÿ|
ÿÿÿvar(_cons)|ÿÿÿ.7369954ÿÿÿ.4889082ÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿ.2008143ÿÿÿÿ2.704799
------------------------------------------------------------------------------
LRÿtestÿvs.ÿlogisticÿmodel:ÿchibar2(01)ÿ=ÿ3.78ÿÿÿÿÿÿÿÿProbÿ>=ÿchibar2ÿ=ÿ0.0259

.ÿ
.ÿ//ÿCheckÿyourÿupdateÿstatus
.ÿquietlyÿqueryÿupdate

.ÿforeachÿcomponentÿinÿexeÿadoÿutilitiesÿdocsÿ{
ÿÿ2.ÿÿÿÿÿassertÿr(inst_`component')ÿ==ÿr(avbl_`component')
ÿÿ3.ÿ}

.ÿ
.ÿexit

endÿofÿdo-file


.

Thanks. It seems the issue is with Stata not up-to-date. The last update on my Stata was in Mar 2020 and this method was introduced after that. I have asked admin to update the Stata and hopefully, it will work.