Python Pandas Fisher Exact Test 2×2 Python Pandas Fisher精确检验2×2

huangapple
go评论104阅读模式
英文:

Python Pandas Fisher Exact Test 2x2

问题

以下是您要翻译的内容：

  1. import pandas as pd
  2. import numpy as np
  4. arr = {
  5.        "A" :[75.3,85.2,90.1,80.4,70.9],
  6.        "B" :[80.6,90.2,85.7,70.8,95.1],
  7.        "C" :[85.2,95.6,80.8,90.4,75.1],
  8.        "D" :[90.3,85.5,80.0,70.5,95.2]
  9.               }
  11. df = pd.DataFrame(arr)

我想在每一行上运行2x2费舍尔精确检验，并将p值保存在新列中。换句话说，我想对以下数组运行2x2费舍尔精确检验：

pvalue = [[df["A"][i], df["B"][i]], [df["C"][i], df["D"][i]]]

对于所有i在0到5之间，并将列表保存为数据框的另一列。

在"scipy.stats"中有一个名为"fisher_exact()"的函数，它将输出2x2数组的奇数值和p值。我目前有以下代码，它可以正常工作：

  1. import scipy.stats as stats
  2. pvalue = []
  4. for i in df.index.values:
  5.     a = df["A"][i]
  6.     b = df["B"][i]
  7.     c = df["C"][i]
  8.     d = df["D"][i]
  9.     
  10.     data = [[a,b],[c,d]]
  11.     
  12.     odd_value, p_value = stats.fisher_exact(data)
  13.     pvalue.append(p_value)
  14.     
  15. df["p_value"] = np.array(pvalue)

但是，对于更大的数组，这需要一些时间。有没有办法使用numpy向量操作来运行这个测试？在这里使用的scipy.stats.fisher_exact()是否有更好的函数？

英文:

I have a pandas dataframe with four columns:

  1. import pandas as pd
  2. import numpy as np
  4. arr = {
  5.        "A" :[75.3,85.2,90.1,80.4,70.9],
  6.        "B" :[80.6,90.2,85.7,70.8,95.1],
  7.        "C" :[85.2,95.6,80.8,90.4,75.1],
  8.        "D" :[90.3,85.5,80.0,70.5,95.2]
  9.               }
  11. df = pd.DataFrame(arr)

I want to run the 2x2 fisher exact test on every row and save the p value on a new column. In other words, I want to run the 2x2 fisher exact test on the array below

pvalue = [[df["A"][i], df["B"][i]], [df["C"][i], df["D"][i]]]

for all i between 0 and 5, and save the list as another column in the dataframe.

There is a function in "scipy.stats" called "fisher_exact()" which will output the odd value and p value of the 2x2 array. I have the following code right now which works:

  1. import scipy.stats as stats
  2. pvalue = []
  4. for i in df.index.values:
  5.     a = df["A"][i]
  6.     b = df["B"][i]
  7.     c = df["C"][i]
  8.     d = df["D"][i]
  9.     
  10.     data = [[a,b],[c,d]]
  11.     
  12.     odd_value, p_value = stats.fisher_exact(data)
  13.     pvalue.append(p_value)
  14.     
  15. df["p_value"] = np.array(pvalue)

However, for larger arrays, this takes a while. Is there a way to run this test using numpy vector operations? Is there a better function than scipy.stats.fisher_exact() to use here?

答案1

得分: 1

  1. from scipy.stats import fisher_exact
  2. import numpy as np
  4. # 使用 fisher_exact 函数计算 p 值，并将结果存储在 NumPy 数组中
  5. np.array([fisher_exact(i)[1] for i in df.values.reshape(-1, 2, 2)])
  7. # 结果如下：
  8. array([1.        , 0.45814062, 0.82723738, 0.6484196 , 0.82550328])
  1. # 你也可以运行以下代码来获得相同的结果：
  3. np.apply_along_axis(lambda x: fisher_exact(x.reshape(-1, 2)), 1, df.values)
  5. # 结果如下，第二列即 `[:, 1]` 是 p 值，第一列是比值比（odds ratios）：
  6. array([[0.99264706, 1.        ],
  7.        [0.84502924, 0.45814062],
  8.        [1.05882353, 0.82723738],
  9.        [0.88888889, 0.6484196 ],
  10.        [0.93333333, 0.82550328]])
英文: 
  1. from scipy.stats import fisher_exact
  2. import numpy as np
  4. np.array([fisher_exact(i)[1] for i in df.values.reshape(-1,2,2)])
  6. array([1.        , 0.45814062, 0.82723738, 0.6484196 , 0.82550328])

You could also run:

  1. np.apply_along_axis(lambda x:fisher_exact(x.reshape(-1,2)), 1, df.values) 
  2. array([[0.99264706, 1.        ],
  3.        [0.84502924, 0.45814062],
  4.        [1.05882353, 0.82723738],
  5.        [0.88888889, 0.6484196 ],
  6.        [0.93333333, 0.82550328]])

Where the second column ie [:, 1] is the p_values and the first column is the odd ratios

答案2

得分: 0

另一个选择是使用 fast-fisher 库。

当运行 fast_fisher_exact_compatibility 而不是 fisher_exact 时，我们可以得到以下结果：

  1. import pandas as pd
  2. import numpy as np
  3. from fast_fisher import fast_fisher_exact_compatibility
  5. >>> arr = {
  6. >>>        "A" :[75.3,85.2,90.1,80.4,70.9],
  7. >>>        "B" :[80.6,90.2,85.7,70.8,95.1],
  8. >>>        "C" :[85.2,95.6,80.8,90.4,75.1],
  9. >>>        "D" :[90.3,85.5,80.0,70.5,95.2]
  10. >>>               }
  12. >>> df = pd.DataFrame(arr)
  14. >>> for i in df.index.values:
  15. >>>     a = df["A"][i]
  16. >>>     b = df["B"][i]
  17. >>>     c = df["C"][i]
  18. >>>     d = df["D"][i]
  20. >>>     data = [[a,b],[c,d]]
  21. >>>     odds_f, pval_f = fast_fisher_exact_compatibility(data)
  23. >>> pval_f
  24. 0.825503275434932

你可以将此作为起点，看看在运行更大的数组时性能是否会提升。

英文:

Another option that you could use is the fast-fisher library.

When running fast_fisher_exact_compatibility instead of fisher_exact, we can yield the following:

  1. import pandas as pd
  2. import numpy as np
  3. from fast_fisher import fast_fisher_exact_compatibility
  5. >>> arr = {
  6. >>>        "A" :[75.3,85.2,90.1,80.4,70.9],
  7. >>>        "B" :[80.6,90.2,85.7,70.8,95.1],
  8. >>>        "C" :[85.2,95.6,80.8,90.4,75.1],
  9. >>>        "D" :[90.3,85.5,80.0,70.5,95.2]
  10. >>>               }
  12. >>> df = pd.DataFrame(arr)
  14. >>> for i in df.index.values:
  15. >>>     a = df["A"][i]
  16. >>>     b = df["B"][i]
  17. >>>     c = df["C"][i]
  18. >>>     d = df["D"][i]
  19.     
  20. >>>     data = [[a,b],[c,d]]
  21. >>>     odds_f, pval_f = fast_fisher_exact_compatibility(data)
  23. >>> pval_f
  24. 0.825503275434932

You could use this as a starting point and see whether performance improves when running on larger arrays.

huangapple
  • 本文由 发表于 2023年7月7日 01:06:59
  • 转载请务必保留本文链接：https://go.coder-hub.com/76631106.html
  • numpy
  • pandas
  • python
  • scipy.stats
匿名

发表评论

匿名网友

:?: :razz: :sad: :evil: :!: :smile: :oops: :grin: :eek: :shock: :???: :cool: :lol: :mad: :twisted: :roll: :wink: :idea: :arrow: :neutral: :cry: :mrgreen:

确定