I would like to use a double hurdle (full) to analyze participation and intensity decisions in stata, but I do not know how to implement this model. However, I have tried to run the regression using craggit command, but craggit specification is collapsing before an optimal solution is attained. What could be the problem? Can full double hurdle solve this problem?
Pl you have singularity problem in your data try to solve first then proceed
@Mohamed, i have checked my dataset and i have no singularity problem. However, there are missing values. Could they be the source of problem? If i choose to drop /delete case wise, i will lose most of observation and neither can i delete variables that are seriously affected. Someone suggested that i replace missing values with the means of the available values within a variable. What will be the likely effect on the results?
Finally, i have cracked the puzzle!
May be you could be going through this: The craggit failed to run because in when i logged quantity sold (the dependent variable in second tier) all observations with zero (0) sales were transformed into missing (.) hence my data was predicted perfectly.
Solution was to recode all missing observations into zero (0) as long they harvested.
Hi Eric
I wanna implement the two stage double hurdle model for my data set. In doing so, I am very confused with some commands dhurdle and craggit.
It's straight forward why I chosen is, due to handle some missing values, true zero value, protest and positive values.
Thank you!
@Eric, Could you share me the stages you passed?
Can I implement double hurdle model as Hickman two stage (In the first stage decision to participation, and then, in the second stage the amount to be paid).
Waiting your feedback patiently.
best,
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Applied Statistics
- Survey Design
- Questionnaire