样本量

sklearn实战-乳腺癌细胞数据挖掘(博客主亲自录制视频教程)

https://study.163.com/course/introduction.htm?courseId=1005269003&utm_campaign=commission&utm_source=cp-400000000398149&utm_medium=share

根据power，effect size,a,决定样本量

# -*- coding: utf-8 -*-
"""
sample size VS effect size VS power
Created on Fri Apr 28 11:00:22 2017@author: toby
"""from statsmodels.stats import powernobs = power.tt_ind_solve_power(effect_size = 0.5, alpha =0.05, power=0.8 )print (nobs)
'''
63.76561177540974
'''effect_size = power.tt_ind_solve_power(alpha =0.05, power=0.8, nobs1=25 )
print(effect_size)
'''
0.8087077886680407
'''

t独立检验中，敏感性（power功效）越高，要求的样本量越大，effect size效应量0.5表示中等效应，如果效应太低，即使显著性<0.05,实验无意义

更好的样本计算脚本来自GitHub

https://github.com/thomas-haslwanter/statsintro_python/tree/master/ISP/Code_Quantlets/07_CheckNormality_CalcSamplesize/sampleSize

# -*- coding: utf-8 -*-
"""
Created on Fri Apr 28 11:12:01 2017@author: toby
"""'''Calculate the sample size for experiments, for normally distributed groups, for:
- Experiments with one single group
- Comparing two groups
'''# Copyright(c) 2015, Thomas Haslwanter. All rights reserved, under the CC BY-SA 4.0 International License# Import standard packages
import numpy as np# additional packages
from scipy.stats import normdef sampleSize_oneGroup(d, alpha=0.05, beta=0.2, sigma=1):'''Sample size for a single group. The formula corresponds to Eq 6.2 in the book.'''n = np.round((norm.ppf(1-alpha/2.) + norm.ppf(1-beta))**2 * sigma**2 / d**2)print(('In order to detect a change of {0} in a group with an SD of {1},'.format(d, sigma)))print(('with significance {0} and test-power {1}, you need at least {2:d} subjects.'.format(alpha, 100*(1-beta), int(n))))return ndef sampleSize_twoGroups(D, alpha=0.05, beta=0.2, sigma1=1, sigma2=1):'''Sample size for two groups. The formula corresponds to Eq 6.4 in the book.'''n = np.round((norm.ppf(1-alpha/2.) + norm.ppf(1-beta))**2 * (sigma1**2 + sigma2**2) / D**2)print(('In order to detect a change of {0} between groups with an SD of {1} and {2},'.format(D, sigma1, sigma2)))print(('with significance {0} and test-power {1}, you need in each group at least {2:d} subjects.'.format(alpha, 100*(1-beta), int(n))))return nif __name__ == '__main__':sampleSize_oneGroup(0.5)print('\n')sampleSize_twoGroups(0.4, sigma1=0.6, sigma2=0.6)