12 Example constructing and interpreting a confidence interval for p#

%%html
<iframe width="700" height="400" src="https://www.youtube.com/embed/sjG9Ns_8x6c/" frameborder="0" allowfullscreen></iframe>

import numpy as np
import pandas as pd
from pandas import Series, DataFrame
import matplotlib.pyplot as plt
import seaborn as sns
from scipy import stats, special

khanacademy

Example constructing and interpreting a confidence interval for p fig 1

x = np.array(20 * [1] + (50 - 20)* [0])

xbar = x.mean()

s = x.std(ddof=1) / np.sqrt(len(x))

z = stats.norm.pdf(0.99) # need improvement

xbar - z * s, xbar + z * s

(0.38289623834908926, 0.4171037616509108)

stats.norm.interval(0.99, loc=xbar, scale=s)

(0.21972950119418946, 0.5802704988058106)

(0.40 - 2.576 * np.sqrt(0.40*0.60/50), 0.40 + 2.576 * np.sqrt(0.40*0.60/50))

(0.2215294847881029, 0.5784705152118972)