From 44fc64645aca8446a96206f5ca876efb92590e9c Mon Sep 17 00:00:00 2001
From: Sergio Correia <sergio.correia@gmail.com>
Date: Mon, 25 Sep 2017 11:41:00 -0400
Subject: [PATCH] Fix bug wtih degrees-of-freedom

If the VCE option was specified as `vce(cluster var1#var2)`, reghdfe was
comparing whether the fixed effects were nested within `var1` instead of
within `var1#var2`.

This reduced SEs and overstated statistical confidence.
---
 src/reghdfe_class.mata |  2 +-
 test/bug_cluster.do    | 21 +++++++++++++++++++++
 test/testall.do        |  2 ++
 3 files changed, 24 insertions(+), 1 deletion(-)
 create mode 100644 test/bug_cluster.do

diff --git a/src/reghdfe_class.mata b/src/reghdfe_class.mata
index ee815f2..d4ed5db 100644
--- a/src/reghdfe_class.mata
+++ b/src/reghdfe_class.mata
@@ -610,7 +610,7 @@ class FixedEffects
 
                 if (cluster_data == .) {
                     if (strpos(clustervar, "#")) {
-                        clustervar = base_clustervars[i_cluster]
+                        clustervar = subinstr(clustervars[i_cluster], "#", " ", .)
                         F = factor(clustervar, sample, ., "", 0, 0, ., 0)
                         cluster_data = F.levels
                         F = Factor() // clear
diff --git a/test/bug_cluster.do b/test/bug_cluster.do
new file mode 100644
index 0000000..887da18
--- /dev/null
+++ b/test/bug_cluster.do
@@ -0,0 +1,21 @@
+noi cscript "reghdfe: prevent bug regression with cluster VCE" adofile reghdfe
+
+* SEE:
+* https://www.statalist.org/forums/forum/general-stata-discussion/general/1409974-reghdfe-4-x-standard-errors-and-categorical-variables
+
+* Dataset
+	sysuse auto, clear
+	egen int turn_foreign = group(turn foreign)
+
+* [TEST] Cluster
+
+	* 1. Run benchmark
+	reghdfe price weight, a(turn#trunk foreign) vce(cluster turn_foreign)
+	storedresults save benchmark e()
+	
+	* 2. Run reghdfe with #
+	reghdfe price weight, a(turn#trunk foreign) vce(cluster turn#foreign)
+	storedresults compare benchmark e(), tol(1e-12) exclude(macro: clustvar1 clustvar cmdline)
+	storedresults drop benchmark
+
+exit
diff --git a/test/testall.do b/test/testall.do
index 2046d7d..2d960b5 100644
--- a/test/testall.do
+++ b/test/testall.do
@@ -54,6 +54,8 @@
 	run inconsistent-r2
 	run singleton-and-fweights
 
+	run bug_cluster
+
 	* -prune- corner cases
 	* run prune