WTP econometric models

[Julian Zuñiga]

Hi statalist users,

I'm trying to estimate WTP values for three different types of public services, ( water supply, aqueduct and waste management ) . So my first thought is use a multinomial logit model (no order presence, and no correlation asumptions) but I have questions:

1. Does the model is right for this kind of sets, or estimate a different model for each one? / these because all three services will be pay in on bill (sum of prices or values).
2. Can I use different independent variables for each public service equation on the multinomial logit?
3. How can I estimate the mean WTP for each one?

Thanks for any kind of help that you can provide me,

All the best.

Julíán

[Carlo Lazzaro]

Julian.
the most prolific Stata expert on this topic is Arne Risa Hole.
Hoping that Arne will chime in; in the meantime googling with the following string: -arne risa hole university of Sheffield- gives back some promising entries.

[Anat Tchetchik]

Julian, are your data based on discrete choice experiment, i.e. where respondents had to choose one option out of a choice set? If so, I believe you you can employ the multinomial logit model with alternative specific constants (ASC) (if you have J alternatives, then you should include J-1 ASCs) and I don't see a problem in using different dependent variables as long as you have price as a common variable. To calc. the WTP for a specific attribute, divide its coefficient by the minus value of the cost coefficient. Said that, I hope that Arne Risa Hole will chime... .

[Julian Zuñiga]

Hello, Carlo and Anatmanes,

Thanks for your comments, I hope to get in contact Arne Risa Hole, I only have two months to finish my paper. hehe

Anatmanes, It's not based on choice experiment it's a contingent valuation-WTP on those three public services with:
1. Dichotomous question WTP, with starting price (estimated by some technicall researchs)
2. WTP max question
3. Follow-up questions.

So there are diferent sets of independent variables (demographic, level of actual provision, etc) for each one of the dependent, including the price (WTP max of the respondents).

Again, thanks for your help and time,

Regards.

[Anat Tchetchik]

Well in this case I'm not sure that the MNL is your model.. Good luck!

[Matthew J. Baker]

Julian Zuñiga

I don't know if it of help, but my package bayesmixedlogitwtp (ssc install bayesmixedlogitwtp) fits a mixed logit model via bayesian means in wtp space. I basically followed Train's approach to normalizing estimated coefficients so that results can be interpreted as wtp's. Here is a description.

Hope that helps!

Matt

[Julian Zuñiga]

Thanks Matt
I'll take a look. Bayesian models are new to me. So I'll kepp in touch.

Julián