变量估计 |设置 1
可变性:衡量数据变化的导入维度,即数据是分散的还是紧密聚集的。也称为分散在机器学习或数据科学中处理数据集时,它涉及许多步骤——方差测量、减少、区分随机可变性和真实可变性。识别实际可变性的来源,根据它做出关于预处理选择或模型选择的决策。
与变异性度量相关的术语:
-> Deviation
-> Variance
-> Standard Deviation
-> Mean Absolute Deviation
-> Meadian Absolute Deviation
-> Order Statistics
-> Range
-> Percentile
-> Inter-quartile Range
- 偏差:我们也可以称之为错误或残差。它是衡量值与中心值/观察值之间的差异/分散程度的度量。
例子 :
Sequence : [2, 3, 5, 6, 7, 9]
Suppose, Central/Observed Value = 7
Deviation = [-5, -4, -2, -1, 0, 2]
- 方差(s 2 ):这是估计方差的最知名的度量,因为它是平方偏差。可以将其称为均方误差,因为它是标准偏差的平均值。
例子 :
Sequence : [2, 3, 5, 6, 7, 9]
Mean = 5.33
Total Terms, n = 6
Squared Deviation = (2 - 5.33)2 + (3 - 5.33)2 + (5 - 5.33)2
(6 - 5.33)2 + (7 - 5.33)2 + (9 - 5.33)2
Variance = Squared Deviation / n
代码 -
Python3
# Variance
import numpy as np
Sequence = [2, 3, 5, 6, 7, 9]
var = np.var(Sequence)
print("Variance : ", var)
Python3
# Standard Deviation
import numpy as np
Sequence = [2, 3, 5, 6, 7, 9]
std = np.std(Sequence)
print("Standard Deviation : ", std)
Python3
# Mean Absolute Deviation
import numpy as np
def mad(data):
return np.mean(np.absolute(
data - np.mean(data)))
Sequence = [2, 4, 6, 8]
print ("Mean Absolute Deviation : ", mad(Sequence))
输出 :
Variance : 5.5555555555555545
- 标准偏差:它是方差的平方根。也称为欧几里得范数。
例子 :
Sequence : [2, 3, 5, 6, 7, 9]
Mean = 5.33
Total Terms, n = 6
Squared Deviation = (2 - 5.33)2 + (3 - 5.33)2 + (5 - 5.33)2
(6 - 5.33)2 + (7 - 5.33)2 + (9 - 5.33)2
Variance = Squared Deviation / n
Standard Deviation = (Variance)1/2
代码 -
Python3
# Standard Deviation
import numpy as np
Sequence = [2, 3, 5, 6, 7, 9]
std = np.std(Sequence)
print("Standard Deviation : ", std)
输出 :
Standard Deviation : 2.357022603955158
- 平均绝对偏差:可以估计这些偏差的典型估计值。如果我们平均这些值,负偏差将抵消正偏差。此外,与平均值的偏差总和始终为零。因此,采用平均偏差本身是一种简单的方法。
例子 :
Sequence : [2, 4, 6, 8]
Mean = 5
Deviation around mean = [-3, -1, 1, 3]
Mean Absolute Deviation = (3 + 1 + 1 + 3)/ 4
Python3
# Mean Absolute Deviation
import numpy as np
def mad(data):
return np.mean(np.absolute(
data - np.mean(data)))
Sequence = [2, 4, 6, 8]
print ("Mean Absolute Deviation : ", mad(Sequence))
输出 :
Mean Absolute Deviation : 2.0