Negative R2?

[azev77]

I'm getting negative values for R2 and R2 Adjusted inconsistent w/ your 1st example.

using DataFrames, RDatasets, FixedEffectModels
df = dataset("plm", "Cigar")
reg(df, @formula(Sales ~ NDI + fe(State) + fe(Year)), Vcov.cluster(:State), weights = :Pop)

                         Fixed Effect Model
====================================================================
Number of obs:               1380  Degrees of freedom:          1335
R2:                       -70.104  R2 Adjusted:              -71.739
F Statistic:              13.3382  p-value:                    0.001
R2 within:                  0.139  Iterations:                     6
Converged:                   true
====================================================================
        Estimate  Std.Error  t value Pr(>|t|)  Lower 95%   Upper 95%
--------------------------------------------------------------------
NDI  -0.00526264 0.00144097 -3.65216    0.001 -0.0081649 -0.00236038
====================================================================

I have the latest version of FixedEffectModels.jl installed.

[eloualiche]

Could you do update FixedEffectModels to the master branch as this has been fixed recently. I also notice your DoF is way off so an update might fix this.
I just ran the readme code and found:

=====================================================================
Number of obs:               1380   Degrees of freedom:            32
R2:                         0.803   R2 Adjusted:                0.798
F Statistic:              13.3382   p-value:                    0.001
R2 within:                  0.139   Iterations:                     6
Converged:                   true
=====================================================================
        Estimate  Std.Error  t value Pr(>|t|)   Lower 95%   Upper 95%
---------------------------------------------------------------------
NDI  -0.00526264 0.00144097 -3.65216    0.000 -0.00808942 -0.00243586
=====================================================================

Try the following:

] add FixedEffectModels#Master
using DataFrames, RDatasets, FixedEffectModels
df = dataset("plm", "Cigar")
reg(df, @formula(Sales ~ NDI + fe(State) + fe(Year)), Vcov.cluster(:State), weights = :Pop)

[azev77]

It didn't work @ first when I tried
] add FixedEffectModels#Master
Then I manually deleted the packages and reinstalled from Master & it worked