Question: Predicted Probability Logit Estimation Stata

[Kevin Musoni]

Hello,
i am a beginner in Stata and i ran a logit regression on a binary dependent variable with binary independent variables.
To get my predicted probabilities, i used the command predict prob, after running my logit regression. Now i do not know which specific command i should use, to display the predicted probabilites. Using "tab prob" I only get a list of numbers that do not tell me something specific.

If you need further information or a screenshot, please just come up to me!

thanks in advance and please excuse the inconvenience of my basic question.

Kevin

[Emad Shehata]

logit y x1 x2
predict yh

[Rich Goldstein]

what do you mean by "display the probabilities"? you don't say how many binary independent variables you have, but here is one way assuming you have only 2:

ta var1 var2, su(prob)

if you have more than 2, then maybe you want to collapse or contract your data so that there is only one observation per combination of the various variables (you can get to "the same place" in many other ways; e.g., use "egen" with the "group" function (and probably the label option), sort by your new variable and list one example of each combination along with the probability; or maybe you want some from of graph??

what you did (using tab) will give you one row for each distinct predicted probability - it appears that that is not what you want; please clarify what you do want to get clearer advice

[Richard Williams]

It is not clear what you want to do with the predicted probabilities once you have them. Maybe something like

logit y x1 x2 x3
predict prob
list y x1 x2 x3 prob

This would give you a listing of the observed variables and the predicted probability of success. But personally, I'd be more interested in, say, the predicted values for some specific sets of x values. Check out the margins command.

[Kevin Musoni]

I do have an dependent variable that is binary called edu_lowersec, and i run a logit regression with 15 independent binary variables. I calculated the marginal effects using the margins,dydx() command. what i need now is the predicted probability of these values.
For example i have the independent variables: mother_edu_dropout, mother_edu_lowersec, mother_edu_secondary, mother_edu_uppersec. Using the marginal effect, I decided to use "edu_lowersec" as my baseline specification, meaning i compare this to the other the variables to get the marginal effect.
What I need now, is the predicted probability, meaning the "real" value of the probability of the 4 variables. Do you know what i mean? If yes, i would love to apreciate your support!

[Kevin Musoni]

Look in the attached document.
So i already calculated the values from the above table. What i need now is the values of the lower one. If understandable, do you know how to get into them? By using which commands?

Note: instead of parental educ i use the education of the mother - see the post above.

[Kevin Musoni]

Again, i like to refresh my question, since i do not know how to solve the lower part of the table..anyone an idea ?