Use Welch's t-test to calculate confidence scores in the bisect script.

tools/auto_bisect/math_utils.py

 # Copyright 2014 The Chromium Authors. All rights reserved.
 # Use of this source code is governed by a BSD-style license that can be
 # found in the LICENSE file.
 
 """General statistical or mathematical functions."""
 
 import math
 
 
 def TruncatedMean(data_set, truncate_percent):
@@ skipping 39 matching lines @@
                           data_set) / kept_weight
 
   return truncated_mean
 
 
 def Mean(values):
   """Calculates the arithmetic mean of a list of values."""
   return TruncatedMean(values, 0.0)
 
 
+def Variance(values):
+  """Calculates the sample variance."""
+  if len(values) == 1:
+    return 0.0
+  mean = Mean(values)
+  differences_from_mean = [float(x) - mean for x in values]
+  squared_differences = [float(x * x) for x in differences_from_mean]
+  variance = sum(squared_differences) / (len(values) - 1)
+  return variance
+
+
 def StandardDeviation(values):
   """Calculates the sample standard deviation of the given list of values."""
-  if len(values) == 1:
-    return 0.0
-
-  mean = Mean(values)
-  differences_from_mean = [float(x) - mean for x in values]
-  squared_differences = [float(x * x) for x in differences_from_mean]
-  variance = sum(squared_differences) / (len(values) - 1)
-  std_dev = math.sqrt(variance)
-
-  return std_dev
+  return math.sqrt(Variance(values))
 
 
 def RelativeChange(before, after):
   """Returns the relative change of before and after, relative to before.
 
   There are several different ways to define relative difference between
   two numbers; sometimes it is defined as relative to the smaller number,
   or to the mean of the two numbers. This version returns the difference
   relative to the first of the two numbers.
 
@@ skipping 39 matching lines @@
 
 
 # Redefining built-in 'StandardError'
 # pylint: disable=W0622
 def StandardError(values):
   """Calculates the standard error of a list of values."""
   if len(values) <= 1:
     return 0.0
   std_dev = StandardDeviation(values)
   return std_dev / math.sqrt(len(values))