include socioeconomic attributes in nested logit

[Linfeng Gong]

Dear Professor Michel,

I'm trying to model a traffic choice modeling using Biogeme. The model I want to build is the nested logit model.

Below is an example of the dataset I used: 

There is a socioeconomic variable named HINC (which stands for household income). I want to include this variable in the nested logic model and assume it exerts different impacts on each alternative. I was wondering is below the right way to include HINC in the utility function?

Another question is: what is the meaning parameter MU? Is this the logsum parameter?

Thank you,

-Linfeng.

[Bierlaire Michel]

The use of socio-economic variables is exactly the same for logit and nested logit. No difference. 

So, yes, your specification looks right.

Regarding MU, are you referring to the online example? http://biogeme.epfl.ch/examples/swissmetro/09nested.py

If so, it is the nest parameter. You need to associate one such parameter with each nest that contains more than one alternative. And the parameter must be larger or equal to one.

On 12 Oct 2021, at 05:53, Linfeng Gong <glf...@gmail.com> wrote:

Dear Professor Michel,

I'm trying to model a traffic choice modeling using Biogeme. The model I want to build is the nested logit model.

Below is an example of the dataset I used: 

<data.PNG>

There is a socioeconomic variable named HINC (which stands for household income). I want to include this variable in the nested logic model and assume it exerts different impacts on each alternative. I was wondering is below the right way to include HINC in the utility function?

<util_fuc.PNG>

Another question is: what is the meaning parameter MU? Is this the logsum parameter?

Thank you,

-Linfeng.

--
You received this message because you are subscribed to the Google Groups "Biogeme" group.
To unsubscribe from this group and stop receiving emails from it, send an email to biogeme+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/biogeme/51532e06-85c3-44b0-aa43-67014b5031aen%40googlegroups.com.
<data.PNG><util_fuc.PNG>

[Linfeng Gong]

Thank you professor. And yes, regarding MU, I’m referring to the online example. I was also wondering is there any online documentation which shows how MU was used when defining the utility function of a nested logit model in Biogeme? Would really appreciate it if you can share me the link.

Thanks again,

-Linfeng.