实验设计：功效分析

Python 中的 A/B 测试

Moe Lotfy, PhD

Principal Data Science Manager

效应量

均值差异的 Cohen's d
比例差异的 Cohen's h

Cohen's h 效应量公式

经验法则
- 小效应 = 0.2
- 中等效应 = 0.5
- 大效应 = 0.8

# Calculate standardized effect size
from statsmodels.stats.proportion import proportion_effectsize
effect_size_std = proportion_effectsize(.33, .3)
print(effect_size_std)

0.0645

# Calculate standardized effect size
from statsmodels.stats.proportion import proportion_effectsize
effect_size_std = proportion_effectsize(p_B, p_A)
print(effect_size_std)

0.0716

比例的样本量估计

# Import power module
from statsmodels.stats import power
# Calculate sample size
sample_size = power.TTestIndPower().solve_power(effect_size=effect_size_std,
                                                power=.80,
                                                alpha=.05,
                                                nobs1=None)
print(sample_size)

3057.547

样本量与 MDE 对功效的影响

# Import t-test power package
from statsmodels.stats.power import TTestIndPower
# Specify parameters for power analysis
sample_sizes = array(range(10, 120))
effect_sizes = array([0.2, 0.5, 0.8])
# Plot power curves
TTestIndPower().plot_power(nobs=sample_sizes, effect_size=effect_sizes)
plt.show()

功效曲线：样本量如何随功效和效应量变化

均值的样本量估计

# Calculate the baseline mean order value
mean_A = checkout[checkout['checkout_page']=='A']['order_value'].mean()
print(mean_A)

24.9564

std_A = checkout[checkout['checkout_page']=='A']['order_value'].std()
print(std_A)

2.418

# Specify the desired minimum average order value
mean_new = 26

# Calculate the standardized effect size
std_effect_size=(mean_new-mean_A)/std_A

均值的样本量估计

sample_size = power.TTestIndPower().solve_power(effect_size=std_effect_size,
                                                power=.80,
                                                alpha=.05,
                                                nobs1=None)
print(sample_size)

85.306

Passons à la pratique !

Python 中的 A/B 测试