Sequential logit

[zakaria]

I'm trying to use Maarten Buis's -seqlogit- package my commandline for regression looks like this seqlogit rbcontrause i.nbrsexcomp i.wage i.rbnewedu i.bnewpartoccu i.bnewspaper i.btv i.bnewrelig i.V190 i.V102 i.rdchild i.bdechealth i.bwant[fw=biwt], or robust tree(1:2 3, 2:3) ofinterest(rbcontrause) Then I want to use seqlogitdecomp, area at(wage 0 rbnewedu 0 bnewpartoccu 0 bnewspaper 0 btv 0 bnewrelig 1 V190 1 V102 1 rdchild 0 bdechealth 0 bwant 0) too see the impact of the first variable in the model i.e. nbrsexcomp but it is showing error varlist required. Zak

[Nick Cox]

That code is more or less unreadable.

As Jon Skeet wrote pertinently

Don't forget that you're basically asking other people to help you out of the goodness of their heart - it's up to you to do all you can to make that as simple as possible.

http://meta.stackexchange.com/questi...tion-checklist

I'd suggest reposting your question and presenting it as readable code.. Toggle the Editor to Advanced (the button A) and use the # button to allow code setting.

[Steve Samuels]

And, as requested in the FAQ, please change your user-name to your real name (first & last). It's a way of showing respect for one another. Contact the List administrator through the CONTACT US link at the bottom of the page.

[Maarten Buis]

The area option in seqlogitdecomp only makes sense when you have specified the over() option in seqlogit. If you want to see the decomposition without specifying the over() option in seqlogit you need to specify the table option in seqlogitdecomp instead of the area option.

[zakaria]

Thanks my colleague has already discussed it with you. Regards


Zak