I am trying to calculate confidence interval for an A/B test. In this A/B test, I have two samples where I have the number of installs for control and treatment group.

I used confint_proportions_2indep to get the CI. However, I am getting “nan” as result.

Similar question was asked here but when I used confint_proportions_2indep function I got nan as my lower and upper intervals.

Here is my code:

AB_control_cnt = control.sum()           # Control Sign-Up Count
AB_treatment_cnt = experiment.sum()      # Treatment Sign-Up Count
AB_control_size = control.count()        # Control Sample Size 
AB_treatment_size = experiment.count()   # Treatment Sample Size

ci = confint_proportions_2indep(AB_treatment_cnt, AB_treatment_size, 
                                AB_control_cnt, AB_control_size, method=None, compare='diff',
                                alpha=0.05, correction=True)

lower = ci[0]
upper = ci[1]
lower_lift = ci[0] / AB_control_rate
upper_lift = ci[1] / AB_control_rate

I searched similar questions on google and stackoverflow but I could not find an answer. Thank you for your help!