pandas 大熊猫与 numpy 中的不同标准
声明：本页面是StackOverFlow热门问题的中英对照翻译，遵循CC BY-SA 4.0协议，如果您需要使用它，必须同样遵循CC BY-SA许可，注明原文地址和作者信息，同时你必须将它归于原作者（不是我）：StackOverFlow
原文地址: http://stackoverflow.com/questions/24984178/
Warning: these are provided under cc-by-sa 4.0 license. You are free to use/share it, But you must attribute it to the original authors (not me):
StackOverFlow
Different std in pandas vs numpy
提问by Mannaggia
The standard deviation differs between pandas and numpy. Why and which one is the correct one? (the relative difference is 3.5% which should not come from rounding, this is high in my opinion).
pandas 和 numpy 的标准偏差不同。为什么，哪一个是正确的？（相对差异是 3.5%，这不应该来自四舍五入，我认为这是很高的）。
Example
例子
import numpy as np
import pandas as pd
from StringIO import StringIO
a='''0.057411
0.024367
0.021247
-0.001809
-0.010874
-0.035845
0.001663
0.043282
0.004433
-0.007242
0.029294
0.023699
0.049654
0.034422
-0.005380'''
df = pd.read_csv(StringIO(a.strip()), delim_whitespace=True, header=None)
df.std()==np.std(df) # False
df.std() # 0.025801
np.std(df) # 0.024926
(0.024926 - 0.025801) / 0.024926 # 3.5% relative difference
I use these versions:
我使用这些版本：
pandas: '0.14.0' numpy: '1.8.1'
Pandas：'0.14.0' numpy：'1.8.1'
回答by NPE
In a nutshell, neither is "incorrect". Pandas uses the unbiased estimator(N-1in the denominator), whereas Numpy by default does not.
简而言之，两者都不是“不正确的”。Pandas 使用无偏估计器（N-1在分母中），而 Numpy 默认不使用。
To make them behave the same, pass ddof=1to numpy.std().
要使它们的行为相同，请传递ddof=1到numpy.std().
For further discussion, see
有关进一步讨论，请参见
回答by Xuan
For pandasto performed the same as numpy, you can pass in the ddof=0parameter, so df.std(ddof=0).
对于pandasto 执行与 相同numpy，您可以传入ddof=0参数，因此df.std(ddof=0)。
This short video explains quite well why n-1might be preferred for samples. https://www.youtube.com/watch?v=Cn0skMJ2F3c
这个简短的视频很好地解释了为什么n-1可能更喜欢样品。https://www.youtube.com/watch?v=Cn0skMJ2F3c
