WTP Estimates

This forum is for posts covering broader stated choice experimental design issues.

Moderators: Andrew Collins, Michiel Bliemer, johnr

WTP Estimates

Postby JvB » Thu Sep 01, 2022 6:03 am

Hi there,

I have several WTP estimates as follows:

Expression Value Robust s.e. Rob t-ratio (0)
WTP_Hoehe300 3.7706 0.8892 4.24
WTP_Hoehe600 2.5797 0.7642 3.38
WTP_Hoehe900 0.6664 0.5987 1.11
WTP_Zeit12 2.6973 0.7349 3.67
WTP_Zeit42 1.9284 0.6663 2.89
WTP_Zeit72 1.1637 0.6612 1.76

I have an experiment with 1,6% as status quo price level so the WTP for a shift in attribute 2 from Hoeheunb (reference category) to Hoehe300 would be 3,7%.
I am confused because status quo level 1,6% + 3,7% = 5,3% total WTP for the mentioned shift in attribute levels BUT the highest price level that was shown in the experiment was 4,8%.
Can it be that the estimated total WTP of 5,3% is higher that the highest shown price level (4,8%)?

Any advice is highly appreciated.
Thank you very much.
JvB
 
Posts: 37
Joined: Mon Mar 22, 2021 12:17 am

Re: WTP Estimates

Postby Michiel Bliemer » Thu Sep 01, 2022 9:29 am

That is an interesting question. While I do not think that the marginal WTP could exceed the price level, I see no reason why the total WTP could not exceed the highest price level, especially if the profile with the best levels for each attribute does not appear in the experimental design.

Michiel
Michiel Bliemer
 
Posts: 1430
Joined: Tue Mar 31, 2009 4:13 pm

Re: WTP Estimates

Postby JvB » Thu Sep 01, 2022 3:15 pm

Dear Michiel,

thank you for your reply.
The math behind my WTP estimation is:

### WTP with Delta-Method
deltaMethod_settings=list(expression=c(WTP_Hoehe300="-1*(b_HoeheEEE300/b_Beitr)", WTP_Hoehe600="-1*(b_HoeheEEE600/b_Beitr)",
WTP_Hoehe900="-1*(b_HoeheEEE900/b_Beitr)", WTP_Zeit12="-1*(b_ZeitEEE12/b_Beitr)",
WTP_Zeit42="-1*(b_ZeitEEE42/b_Beitr)", WTP_Zeit72="-1*(b_ZeitEEE72/b_Beitr)" ))
apollo_deltaMethod(model, deltaMethod_settings)

With b_Beitr being the price level.
So I think my total WTP estimation should be fine and also the 3,7% MWTP as it relies on the parameters and depending on their size it can be that for the total WTP extrapolation is necessary?!

Thank you very much.
JvB
 
Posts: 37
Joined: Mon Mar 22, 2021 12:17 am

Re: WTP Estimates

Postby Michiel Bliemer » Thu Sep 01, 2022 4:21 pm

I have only used Apollo once so cannot really comment on this R script, but ratios of parameters is correct if the attributes are numerical and not dummy or effects coded. If HoeheEEE300 and HoeheEEE60 are dummy coded coefficients then the WTP needs to be interpreted somewhat differently and is not simply a ratio of parameters. You may want to post your question on the Apollo forum.

Michiel
Michiel Bliemer
 
Posts: 1430
Joined: Tue Mar 31, 2009 4:13 pm


Return to Choice experiments - general

Who is online

Users browsing this forum: No registered users and 4 guests

cron