ksecology/fixedeffectmodel

Fixedeffectmodel: panel data modeling in Python

fixedeffectmodel's Issues

iv2sls() error

Hello,

when I tried to run the sample code I run into the following error.

Is my pandas version not correct for this package to run? Thanks!

How to set up the fixed effect model in python

Fixed Effect Regression does not work

Here, I'm doing everything fine that seems to me, but unfortunately, the package does not work as it should. I'm using the right parameters, Please have a look at the code and the error message in the attached image.

Thanks & regards
Kaleem

summary() not working

Hello, I´ve been trying to perform a iv2sls regression with fixed effects with your library but I can´t get the summary of the results. I even tried it with the data from your example and it doesn´t work.

If you could please check this out I would really appreciate it. It would be really useful to be able to get the results. This seems to be the only library with decent outputs for this kind of regression in Python. Thanks.

'result.summary()' does not work

I have made some exercises of the examples, but I found that the code 'result.summary()' in each example gives some warnings.
Please help me. thx

Storing absorbed fixed effects

Can the absorbed fixed effects be recovered/stored, as in the STATA version of reghdfe?

Error prevents code to run properly

The code seems to be broken when transforming estimates to strings for visualization into table. OLSFixed.py line 213:

        params_data = lzip(["%#6.5f" % float(params[i]) for i in exog_len],
                           ["%#6.5f" % float(std_err[i]) for i in exog_len],
                           ["%#6.4f" % float(tstat[i]) for i in exog_len],
                           ["%#6.4f" % float(prob_stat[i]) for i in exog_len],
                           ["%#6.4f" % float(conf_int[0][i]) for i in exog_len],
                           ["%#6.4f" % float(conf_int[1][i]) for i in exog_len])

I needed to cast the various parameters to float for the code to work again. I am using Python 3.9.6 on MacOS

fixedeffect() error

hello, I have been using fixedeffect function, but there's an error when I include two category variables(userid and time).

It shows 'NameError: Total sum of square equal 0, program quit.'

The function works well when I include only one of these two category variables.
Would you please check this out? thanks a lot.

FixedEffectModel

re: replacing NaN with zeros

Any thoughts on omitting the preprocessing step in which NaNs are set to zero? Choosing how to handle missing data is a non-trivial modeling choice and having it happen automatically might cause problems for the analyst. My preference would be to drop any row with missings, but throwing an error is also a fine way to go. Thanks!

Question: Partialout

Hi
I was looking into the command, and the helpfile. I couldnt find if there is an option to simply obtain the demean versions of the data.
For example, in Stata you could use the user written command hdfe. In Julia you have partialout as part of the package FixedEffectModel. In R, you have demean, part of fixest.
Is there a similar option with this package for python?
Thank you
Fernando