import numpy as np

rng = np.random.default_rng()
num_repetitions = 20000
L = []
for i in np.arange(num_repetitions):
    L.append(rng.choice([0, 1], size=100, replace=True, p=[0.74, 0.26]).sum())

from matplotlib.pyplot import subplots

fig, ax = subplots(figsize = (8, 8))
ax.hist(L, bins = np.arange(0, 50, 1));
ax.scatter(8, 0, color='red', s=100);

import pandas as pd

data = [['Asian/PI', 0.15, 0.26], 
        ['Black/AA', 0.18, 0.08], 
        ['Caucasian', 0.54, 0.54],
        ['Hispanic', 0.12, 0.08],
        ['Other',0.01, 0.04]]

df = pd.DataFrame(data, columns=['Ethnicity', 'Eligible', 'Panels'])
print(df)
df.plot.barh('Ethnicity');

   Ethnicity  Eligible  Panels
0   Asian/PI      0.15    0.26
1   Black/AA      0.18    0.08
2  Caucasian      0.54    0.54
3   Hispanic      0.12    0.08
4      Other      0.01    0.04

x, y = np.unique(rng.choice(df['Ethnicity'], size=1453, replace=True, p=df['Eligible']), return_counts=True)
y = y/1453
print(y)
df.insert(loc=3, column='Random', value=y)
print(df)
df.plot.barh('Ethnicity')

[0.15485203 0.16311081 0.54439092 0.12594632 0.01169993]
   Ethnicity  Eligible  Panels    Random
0   Asian/PI      0.15    0.26  0.154852
1   Black/AA      0.18    0.08  0.163111
2  Caucasian      0.54    0.54  0.544391
3   Hispanic      0.12    0.08  0.125946
4      Other      0.01    0.04  0.011700

<Axes: ylabel='Ethnicity'>

def calculate_tvd(X, Y):
    A = np.abs(X - Y).sum()/2
    return A

num_repetitions = 10000
L=[]
for i in range(num_repetitions):
    x, y = np.unique(rng.choice(df['Ethnicity'], size=1453, replace=True, p=df['Eligible']), return_counts=True)
    y = y/1453
    L.append(calculate_tvd(df['Eligible'], y))

fig, ax = subplots(figsize = (8, 8))
ax.hist(L, bins=np.arange(0.0, 0.2, 0.001));
ax.scatter(0.14, 0, color='red')
    

<matplotlib.collections.PathCollection at 0x104e43cd0>