CODEWITHSUNDEEP
pythonscipystatisticsscipy.stats
cloudycider
cloudycider

Reputation: 37

Calculating p-value with Python. One-tailed or two-tailed?

I'm using this function to calculate the p-value of an experiment I'm running. I'm not sure if it's one-tailed or two-tailed. How can I infer this from the code? Thanks

from scipy import stats

def get_pvalue(con_conv, test_conv, con_size, test_size):  
    lift =  - abs(test_conv - con_conv)

    scale_one = con_conv * (1 - con_conv) * (1 / con_size)
    scale_two = test_conv * (1 - test_conv) * (1 / test_size)
    scale_val = (scale_one + scale_two)**0.5

    p_value = 2 * stats.norm.cdf(lift, loc = 0, scale = scale_val )

    return p_value

Upvotes: 0

Views: 793

Answers (1)

merv
merv

Reputation: 76700

Two-tailed. This can be inferred because the difference uses an absolute value, and the p_value multiplies the (one-sided) tail of the normal CDF by 2.

Upvotes: 1

Related Questions