scipy.stats.binom_test¶

scipy.stats.binom_test(x, n=None, p=0.5, alternative='two-sided')[source]¶

Perform a test that the probability of success is p.

This is an exact, two-sided test of the null hypothesis that the probability of success in a Bernoulli experiment is p.

Parameters:

Parameters:	x : integer or array_like the number of successes, or if x has length 2, it is the number of successes and the number of failures. n : integer the number of trials. This is ignored if x gives both the number of successes and failures p : float, optional The hypothesized probability of success. 0 <= p <= 1. The default value is p = 0.5 alternative : {‘two-sided’, ‘greater’, ‘less’}, optional Indicates the alternative hypothesis. The default value is ‘two-sided’.
Returns:	p-value : float The p-value of the hypothesis test

x : integer or array_like

the number of successes, or if x has length 2, it is the number of successes and the number of failures.

n : integer

the number of trials. This is ignored if x gives both the number of successes and failures

p : float, optional

The hypothesized probability of success. 0 <= p <= 1. The default value is p = 0.5

alternative : {‘two-sided’, ‘greater’, ‘less’}, optional

Indicates the alternative hypothesis. The default value is ‘two-sided’.

Returns:

p-value : float

The p-value of the hypothesis test

References