P value of test should use cumulative distribution function. The probability-density function of a distribution means the density function at a certain point!
For example, if we are testing mu1 < mu2 with t-score of t, we get p-value by cdf(t) = P(X < t) whereas pdf(t) = P(X=t)
Tested the result and now they agree with scipy function test_ind_from_stats
P value of test should use cumulative distribution function. The probability-density function of a distribution means the density function at a certain point!
For example, if we are testing mu1 < mu2 with t-score of t, we get p-value by cdf(t) = P(X < t) whereas pdf(t) = P(X=t)
Tested the result and now they agree with scipy function
test_ind_from_stats