Hello everyone,

Currently I am busy doing my research and I am thinking about running a Tobit regression, however I am not quite sure if the Tobit model is the right model to use.

My data is structered as follows, I have a dependent variable that is naturally ordered and can take on discrete values of 0, 1, 2, 3 and 4. I have two independent dummy variables, one independent variable that can take on all values of zero and above, and an indicator variable for the nation (one of twelve) in which the investment is executed. I have +/- 2000 observations.

Do you think that the tobit model with a lower limit of 0 and upper limit of 4 is the right model to use, or is there some other model that might fit this data better?

Additionally if the tobit model is the right model to use, I am wondering what the /sigma means that the stata output gives me and why this is important.

Thank you all!

Best,
Luke