comparison of WTP values

[Anne Posen]

Dear all,

I wish to compare the Willingness-To-Pay values (in both, preference and WTP-space) between male and female respondents from the same sample. However, I don't know how to do this. I have read about the Poe-Test but do not know how to implement it. As far as I can see, the other entries into this forum with respect to the Poe-test have not been solved. Is there another valid way to compare WTP values in stata?

I am looking forward to your response.

Kind regards and thanks in advance,
Anne

[Jeff Wooldridge]

Anne: Which command are you using to estimate WTP? It’s preferred if you show some output.

[Anne Posen]

Dear Jeff,

thank you very much for your reply. I am using wtp and mixlogitwtp in Stata 15.1.

below you can see a screenshot of my dofile as .png

I wanted to calculate the wtp values for male and female participants respectively and then compare their values via the Poe-Test (or a similar test)

Kind regards,
Anne

[Jeff Wooldridge]

I haven't used these commands and so I'm unsure of the output. Can you show the Stata output? I assume it produces a standard error, probably via the bootstrap.

[Anne Posen]

Of course. I also just noticed that my command for mixlogitwtp is incorrect. If I am not mistaken it should be

mixlogitwtp choice, group(csetfort) price (digital_interest_multi) rand(digital digital_loan digital_distance digital_installment digital_cost) id(Nr) nrep(100)

My results for e.g. the wtp for digital credit by women (estimation in wtp space) :

My results for e.g. the wtp for digital credit by men (estimation in wtp space) :

[Jenna Grabey]

Anne Posen we cannot see from your last post what the error message is. Do you have some text in red to indicate the error? Stata will also provide you with a link to try and help you solve the error. I am aware that this link does not always help as it is provides quite generalised information.

[Anne Posen]

Dear Jenna,

thank you very much for your message.

There is no error code so far. My issue is rather how I proceed with these outputs. I would like to compare the wtp of the female with the wtp of the male output. The literature suggests to use the test by Poe, G. L., Giraud, K. L., & Loomis, J. B. (2005). Yet, I am not sure how to implement this and was wondering, whether there is a stata command for this issue.

Kind regards,
Anne

[big boy]

Dear Anne, have you gotten any help on how to implement the Poe test? If yes, please kindly share the information. I am also trying to implement the test. Jenna Grabey and Arne Risa Hole, please do you have any idea on how to implement the Poe test.

Thanks.

[Paul Arnem]

Dear all,

has anyone found a solution for this? The "poetest" command works in preference space, but what if I want to compare coefficients in WTP space?

Thanks a lot