display both types of dof (#231)

* display both  types of df

* Update README.md

* Update Project.toml

* Update README.md

* Update README.md

Project.toml

@@ -1,6 +1,6 @@
 name = "FixedEffectModels"
 uuid = "9d5cd8c9-2029-5cab-9928-427838db53e3"
-version = "1.9.1"
+version = "1.9.2"
 
 [deps]
 DataFrames = "a93c6f00-e57d-5684-b7b6-d8193f3e46c0"

README.md

@@ -17,17 +17,18 @@ The objective of the package is similar to the Stata command [`reghdfe`](https:/
 using DataFrames, RDatasets, FixedEffectModels
 df = dataset("plm", "Cigar")
 reg(df, @formula(Sales ~ NDI + fe(State) + fe(Year)), Vcov.cluster(:State), weights = :Pop)
-# =====================================================================
-# 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
-# =====================================================================
+                            FixedEffectModel                            
+=========================================================================
+Number of obs:                 1380   Converged:                     true
+dof (model):                      1   dof (residuals):                 45
+R²:                           0.803   R² adjusted:                  0.798
+F-statistic:                13.3382   P-value:                      0.001
+R² within:                    0.139   Iterations:                       5
+=========================================================================
+        Estimate  Std. Error    t-stat  Pr(>|t|)   Lower 95%    Upper 95%
+─────────────────────────────────────────────────────────────────────────
+NDI  -0.00526264  0.00144097  -3.65216    0.0007  -0.0081649  -0.00236038
+=========================================================================
 ```
 
 

src/FixedEffectModel.jl

@@ -24,9 +24,8 @@ struct FixedEffectModel <: RegressionModel
     contrasts::Dict
 
     nobs::Int64             # Number of observations
-    dof::Int64              # Number parameters estimated - has_intercept
-    dof_residual::Int64        # dof used for t-test and F-stat. nobs - degrees of freedoms with simple std
-
+    dof::Int64              # Number parameters estimated - has_intercept. Used for p-value of F-stat.
+    dof_residual::Int64     # dof used for t-test and p-value of F-stat. nobs - degrees of freedoms with simple std
     rss::Float64            # Sum of squared residuals
     tss::Float64            # Total sum of squares
     r2::Float64             # R squared
@@ -36,8 +35,8 @@ struct FixedEffectModel <: RegressionModel
     p::Float64              # p value for the F statistics
 
     # for FE
-    iterations::Union{Int, Nothing}         # Number of iterations
-    converged::Union{Bool, Nothing}         # Has the demeaning algorithm converged?
+    iterations::Int         # Number of iterations
+    converged::Bool         # Has the demeaning algorithm converged?
     r2_within::Union{Float64, Nothing}      # within r2 (with fixed effect
 
     # for IV
@@ -178,7 +177,9 @@ end
 function top(m::FixedEffectModel)
     out = [
             "Number of obs" sprint(show, nobs(m), context = :compact => true);
-            "Degrees of freedom" sprint(show, dof(m), context = :compact => true);
+            "Converged" m.converged;
+            "dof (model)" sprint(show, dof(m), context = :compact => true);
+            "dof (residuals)" sprint(show, dof_residual(m), context = :compact => true);
             "R²" @sprintf("%.3f",r2(m));
             "R² adjusted" @sprintf("%.3f",adjr2(m));
             "F-statistic" sprint(show, m.F, context = :compact => true);

src/fit.jl

@@ -236,8 +236,8 @@ function StatsAPI.fit(::Type{FixedEffectModel},
 
     # compute tss now before potentially demeaning y
     tss_total = tss(y, has_intercept | has_fe_intercept, weights)
-    # create unitilaized
-    iterations, converged, r2_within = nothing, nothing, nothing
+    # initalize fields
+    iterations, converged, r2_within = 0, true, nothing
     F_kp, p_kp = nothing, nothing
     collinear_fe = falses(length(var_names_all))
 
@@ -441,22 +441,20 @@ function StatsAPI.fit(::Type{FixedEffectModel},
             end
         end
     end
-    dof_residual_ = max(1, nobs - size(X, 2) - dof_fes - dof_add)
-    dof_ = max(1, size(X, 2) - (has_intercept | has_fe_intercept))
-
 
     nclusters = nothing
     if vcov isa Vcov.ClusterCovariance
         nclusters = Vcov.nclusters(vcov_method)
     end
 
-
     # Compute standard error
-    vcov_data = Vcov.VcovData(Xhat, crossx, residuals, dof_residual_)
+    vcov_data = Vcov.VcovData(Xhat, crossx, residuals, nobs - size(X, 2) - dof_fes - dof_add)
     matrix_vcov = StatsAPI.vcov(vcov_data, vcov_method)
 
     # Compute Fstat
     F = Fstat(coef, matrix_vcov, has_intercept)
+    # dof_ is the number of estimated coefficients beyond the intercept.
+    dof_ = size(X, 2) - has_intercept
     dof_tstat_ = max(1, Vcov.dof_residual(vcov_data, vcov_method) - has_intercept | has_fe_intercept)
     p = fdistccdf(dof_, dof_tstat_, F)
     # Compute Fstat of First Stage
@@ -477,7 +475,7 @@ function StatsAPI.fit(::Type{FixedEffectModel},
     rss = sum(abs2, residuals)
     mss = tss_total - rss
     r2 = 1 - rss / tss_total
-    adjr2 = 1 - rss / tss_total * (nobs - (has_intercept | has_fe_intercept)) / dof_residual_
+    adjr2 = 1 - rss / tss_total * (nobs - (has_intercept | has_fe_intercept)) / max(nobs - size(X, 2) - dof_fes - dof_add, 1)
     if has_fes
         r2_within = 1 - rss / tss_partial
     end