The margins command calculates predicted probabilities for each covariate across groups (outcome). For example, if you estimate a multinomial model with 3 choices the sum of the predicted probabilities for each covariate will equal one across outcome groups. This is equivalent to running crosstabs with the row option. However, is there an equivalent in margins of running crosstabs with the column option? In other words, is there an option for margins that calculates predicted probabilities that sum to one for each covariate within, as opposed to across, groups.
Related Posts with Predictions after mlogit
sem vs gsem?Hi everyone. I want to use sem or gsem to evaluate the associations of different patterns of weight …
Rounding with a macroHi, To display a number rounded down to a user selected number of decimal places, I wrote some code…
Seeking an efficient way to establish auxiliary time scales refrenced to data elementsI have been working for 2 + years with data generated from a custom app seeking improvement in treat…
How to replace inconsistent identity code with the majority identity within 6 years?I have 140,000 firms for 2010-2015. I want to check whether the industrial code (indcode) is consist…
Generating a variable to count # of "ideal" in each category per observationDear all, I have three variables(idealFruitVeg ; idealfiber ; idealssbweek). Each variable can have…
Subscribe to:
Post Comments (Atom)
0 Response to Predictions after mlogit
Post a Comment