Announcement

Collapse
No announcement yet.
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • Interpretation of hurdle model (upper limit selection)

    Hi,
    I am an undergraduate student, currently analyzing an experimental data (in Public Goods game) for my project. I've been using Tobit estimation but recently got to learn about hurdle model which might be more appropriate in my analysis. The Stata manual gives good explanation http://www.stata.com/manuals14/rchurdle.pdf of this model in general, and as far as I understand, the command: churdle linear will conduct probit estimation for the hurdles and run truncated linear regression for the part the passes those hurdles. however I still am confused about how to interpret the coefficients in upper selection model - If, for example I want to model the amount of punishment points assigned by an individual (The minimum points they can assign is 0 and the maximum is 10) .
    The dependent variable for lower selection model will be the likelihood that subject decides to punish. Would the dependent variable of upper limit selection be the probability that a subject assigns less than 10 points? Or would it be the likelihood of maximum punishment points?

    Thank you in advance,





Working...
X