From 3468661c6f5f12854549a36a124e2884bb471837 Mon Sep 17 00:00:00 2001
From: Damonamajor <56321109+Damonamajor@users.noreply.github.com>
Date: Tue, 16 Sep 2025 10:05:15 -0500
Subject: [PATCH 1/2] Update formulas.R

---
 R/formulas.R | 12 ++++++++++--
 1 file changed, 10 insertions(+), 2 deletions(-)

diff --git a/R/formulas.R b/R/formulas.R
index b78e134..7ea776a 100644
--- a/R/formulas.R
+++ b/R/formulas.R
@@ -178,8 +178,16 @@ prb <- function(assessed, sale_price, na.rm = FALSE) {
 # Calculate the Gini cofficients needed for KI and MKI
 calc_gini <- function(assessed, sale_price) {
   df <- data.frame(av = assessed, sp = sale_price)
-  df <- df[order(df$sp, -df$av), ]
-  assessed_price <- df$av
+  # This Gini coefficient algorithm is sensitive to the order of the input
+  # observations: If multiple observations share the same sale price but have
+  # different estimates, the output coefficients will be different depending
+  # on which of the sales with identical prices gets ordered first in the
+  # input dataframe. To ensure a stable sort order, Quintos recommends
+  # sorting by ascending sale price and then by descending estimate to break
+  # any ties. This produces "worst case" MKI/KI statistics, but ensures those
+  # statistics are deterministic. See this issue for more discussion:
+  # https://github.com/ccao-data/assesspy/issues/33#issuecomment-3180632954
+  df <- df[order(df$sp, -df$av), ]  assessed_price <- df$av
   sale_price <- df$sp
   n <- length(assessed_price)
 

From 33d1bbc4cb3eeb62ba03f85681b825572ecf1f7d Mon Sep 17 00:00:00 2001
From: Damonamajor <56321109+Damonamajor@users.noreply.github.com>
Date: Tue, 16 Sep 2025 10:05:55 -0500
Subject: [PATCH 2/2] Update formulas.R

---
 R/formulas.R | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/R/formulas.R b/R/formulas.R
index 7ea776a..3433dca 100644
--- a/R/formulas.R
+++ b/R/formulas.R
@@ -187,7 +187,8 @@ calc_gini <- function(assessed, sale_price) {
   # any ties. This produces "worst case" MKI/KI statistics, but ensures those
   # statistics are deterministic. See this issue for more discussion:
   # https://github.com/ccao-data/assesspy/issues/33#issuecomment-3180632954
-  df <- df[order(df$sp, -df$av), ]  assessed_price <- df$av
+  df <- df[order(df$sp, -df$av), ]
+  assessed_price <- df$av
   sale_price <- df$sp
   n <- length(assessed_price)