From: Raymond Hettinger <rhettinger@users.noreply.github.com>
Date: Thu, 6 May 2021 14:43:13 +0000 (-0700)
Subject: Eliminate duplicated calculations and unnecessary work for linear regression (GH... 
X-Git-Tag: v3.11.0a1~1193
X-Git-Url: http://git.ipfire.org/gitweb.cgi?a=commitdiff_plain;h=55b78ce3c4e23abe4f27bf16d7968f8851532e47;p=thirdparty%2FPython%2Fcpython.git

Eliminate duplicated calculations and unnecessary work for linear regression (GH-25922)
---

diff --git a/Lib/statistics.py b/Lib/statistics.py
index edb11c868c1c..db8c581068b7 100644
--- a/Lib/statistics.py
+++ b/Lib/statistics.py
@@ -952,11 +952,16 @@ def linear_regression(regressor, dependent_variable, /):
         raise StatisticsError('linear regression requires that both inputs have same number of data points')
     if n < 2:
         raise StatisticsError('linear regression requires at least two data points')
+    x, y = regressor, dependent_variable
+    xbar = fsum(x) / n
+    ybar = fsum(y) / n
+    sxy = fsum((xi - xbar) * (yi - ybar) for xi, yi in zip(x, y))
+    s2x = fsum((xi - xbar) ** 2.0 for xi in x)
     try:
-        slope = covariance(regressor, dependent_variable) / variance(regressor)
+        slope = sxy / s2x
     except ZeroDivisionError:
         raise StatisticsError('regressor is constant')
-    intercept = fmean(dependent_variable) - slope * fmean(regressor)
+    intercept = ybar - slope * xbar
     return LinearRegression(intercept=intercept, slope=slope)