interpretation of results from the fixed effects model with binary variables.

[Karina Ferracioli]

Hello, I need some help with the correct form of interpretation of a fixed effect panel data model using binary variables
The model is:
dependet variable (income), and the independent variables are 5 different regions (A,B,C,D,E) of a country.
Of course that if I use all 5, stata will drop one because of multicollinearity. Then, if I supress region A, what's the correct interpretation for coeficients of variables (B, C, D and E), and how can I find the effect of A on Y?
I want to know if living in B, is better for income than living in C for example.

[Clyde Schechter]

Assuming you used a linear regression, the coefficient of B will be the expected difference in income between region B and region A. The coefficient of C will be the expected difference in income between region C and region A, et c.

You cannot find the effect of A on Y--there is no such thing. Because of the colinearity among the 5 region indicators, their separate effects are undefinable, non-existent. Only the differences among those effects are estimable. You can get the expected income in each region, and you do that with the -margins- command.

If you want to know the expected difference in income between region B and region C, you can either re-run the model with B or C as the reference (omitted) category for region, or you can use the same regression command you did before and then use the -lincom- command to get the difference between their coefficients. -help lincom-

Because you show neither example data nor the actual output from your regression, I cannot give you more concrete advice on the syntax for -margins- and -lincom-. If you can't figure that out after reading -help lincom- and -help margins-, then you can post back showing those things. When showing Stata output, be sure to wrap it in code delimiters so it is properly aligned and easy to read. (If you are not familiar with code delimiters, read the Forum FAQ, with special attention to #12 for instructions.) And when showing example data, be sure to use the -dataex- command. If you are running version 16 or a fully updated version 15.1 or 14.2, -dataex- is already part of your official Stata installation. If not, run -ssc install dataex- to get it. Either way, run -help dataex- to read the simple instructions for using it. -dataex- will save you time; it is easier and quicker than typing out tables. It includes complete information about aspects of the data that are often critical to answering your question but cannot be seen from tabular displays or screenshots. It also makes it possible for those who want to help you to create a faithful representation of your example to try out their code, which in turn makes it more likely that their answer will actually work in your data.

[Marcos Almeida]

Welcome to the Stata Forum / Statalist.

Please take a look at the FAQ. There you'll find information about sharing command/output/data in the Forum

Short answer: use margins followed by marginsplot.

Long answer: to get it (and this is a forecast, not a commitment of mine), please share exactly the command you used, plus the output.

Intermediate answer: when using categorical predictor, one category is taken as reference. There is no need to use binary (dummy) variables for that matter.

[Karina Ferracioli]

Thank you so much for your answers, I just needed the general idea, The answers you gave me are perfect!!!!

[Jan Sadowski]

I am sorry for the redundant entry, I found an error related to a rescaling of the dependent variable when copied from Stata into an Excel spreadsheet; please discard my current post.