Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: tobit, ordered probit, or just OLS???

From   "Augusto Pallavicini" <[email protected]>
To   <[email protected]>
Subject   st: tobit, ordered probit, or just OLS???
Date   Fri, 20 Jul 2012 03:17:28 +0100

Hi all,

I am doing a contingent valuation method about willingness to pay for wind energy based on surveys.
I've got only 69 obs,.

The dependent variable is WTP and the values I obtained are 0 2.5 5 7.5 10 15. The original question from which I obtained the data was.. Would you be willing to pay 2.5 pounds for...? If yes, would you be willing to pay 5...? and so forth... Then I've got independent variables related to socioeconomic factors and energy preferences.

I know I can treat the dv as continuous and use tobit or two part model (i.e. first probit and then truncated), or just OLS. Also, data is censored at ZERO Also, I could use ordered probit or logit but I loose some of the information I have.

Due to the small amount of obs, what model do you recommend me to use? And, how should I decide between the different alternatives?

Thanks in advance

*   For searches and help try:

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index