Support for Microeconometrics about regressiontables.jl HOT 5 CLOSED

Yes, would be great. Ideally, your package would provide a DataFrameRegressionModel, like GLM.jl, then regtable() should be able to use it. Otherwise, you'd need to change regtable() (but I'd prefer the former variant -- otherwise we're duplicating interfaces and RegressionTables would be requiring several estimation packages).

from regressiontables.jl.

Is it absolutely necessary to provide a DataFrameRegressionModel? The current output is a RegressionModel (note that DataFrameRegressionModel is a subtype of RegressionModel). The standard methods are supported (coef, coefnames, nobs, r2, etc.). It may not even be necessary to require Microeconometrics.jl as long as RegressionTables.jl provides support for abstract statistical models.

from regressiontables.jl.

I'd be happy to work with a RegressionModel if it contains all the information we need for the table (notably variable names).

from regressiontables.jl.

One can get covariate labels from coefnames. I don't think StatsModels.jl provides a method to retrieve the name of the dependent variable though. We could propose it in a pull request – it would be a good addition.

from regressiontables.jl.

Closing this as RegressionTables now supports RegressionModels.

from regressiontables.jl.