Independent variables for nested logit model

[pof]

Is possible to work with all independent variables as dicotomics?

[Maguin, Eugene]

Yes, but. The 'but' is the same as for an ordinary logistic regression. In your model you have a categorical, possibly dichotomous, dependent variable (DV) and independents (IVs) that are categorical. You need to make sure that within every level 1 group a) none of the variables (IVs or DV) have a zero frequency cell and b) the crosstabs of IV against DV do not have zero cells. If you have IVs with zero cells, you could, of course, collapse categories for that variable. It might be useful to use split files and run the model using logistic.

Gene Maguin

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of pof
Sent: Tuesday, May 19, 2015 6:50 PM
To: [hidden email]
Subject: Independent variables for nested logit model

Is possible to work with all independent variables as dicotomics?



--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Independent-variables-for-nested-logit-model-tp5729577.html
Sent from the SPSSX Discussion mailing list archive at Nabble.com.

=====================
To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[pof]

Dear Eugene Maguin

thank you for your help.
Is possible to have your private e-mail, because I need help, so is better to contact you directly.
I will wait for your reply.
Best regards,
Paula Odete.

On 20/05/2015 14:23, Maguin, Eugene [via SPSSX Discussion] wrote:

Yes, but. The 'but' is the same as for an ordinary logistic regression. In your model you have a categorical, possibly dichotomous, dependent variable (DV) and independents (IVs) that are categorical. You need to make sure that within every level 1 group a) none of the variables (IVs or DV) have a zero frequency cell and b) the crosstabs of IV against DV do not have zero cells. If you have IVs with zero cells, you could, of course, collapse categories for that variable. It might be useful to use split files and run the model using logistic.

Gene Maguin

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of pof
Sent: Tuesday, May 19, 2015 6:50 PM
To: [hidden email]
Subject: Independent variables for nested logit model

Is possible to work with all independent variables as dicotomics?



--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Independent-variables-for-nested-logit-model-tp5729577.html
Sent from the SPSSX Discussion mailing list archive at Nabble.com.

=====================
To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

---

If you reply to this email, your message will be added to the discussion below:

http://spssx-discussion.1045642.n5.nabble.com/Independent-variables-for-nested-logit-model-tp5729577p5729579.html

To unsubscribe from Independent variables for nested logit model, click here.
NAML

[Art Kendall]

In addition to what Gene suggests:
It is possible to deal with IV's as dichotomies.  It is often undesirable to do.

Is it at all possible fro you to represent your constructs in a less coarse (finer grained) way?

advice would be more firmly grounded if  you explained what your constructs are, how the data were gathered, and how the variables are operationalized (names, labels, values, value labels, etc.)