Sampling Distribution of a Statistic¶

In [1]:

import pandas as pd
import seaborn as sns

import pandas as pd import seaborn as sns

In [2]:

data = pd.read_csv("../data/loans_income.csv")
data

data = pd.read_csv("../data/loans_income.csv") data

Out[2]:

	x
0	67000
1	52000
2	100000
3	78762
4	37041
...	...
49995	40000
49996	54000
49997	50000
49998	82000
49999	70000

50000 rows × 1 columns

In [3]:

sample_data = pd.DataFrame({
    'income': data["x"].sample(1000),
    'type': "Data"
})
sample_data

sample_data = pd.DataFrame({ 'income': data["x"].sample(1000), 'type': "Data" }) sample_data

Out[3]:

	income	type
18869	70000	Data
34424	103000	Data
11601	51985	Data
39244	46000	Data
41527	36000	Data
...	...	...
8679	100000	Data
6003	86100	Data
22853	49898	Data
43955	82000	Data
37150	45000	Data

1000 rows × 2 columns

In [4]:

sample_mean_5 = pd.DataFrame({
    'income': [data["x"].sample(5).mean() for _ in range(1000)],
    'type': "Mean of 5"
})

sample_mean_20 = pd.DataFrame({
    'income': [data["x"].sample(20).mean() for _ in range(1000)],
    'type': "Mean of 20"
})

samples = pd.concat([sample_data, sample_mean_5, sample_mean_20])
samples

sample_mean_5 = pd.DataFrame({ 'income': [data["x"].sample(5).mean() for _ in range(1000)], 'type': "Mean of 5" }) sample_mean_20 = pd.DataFrame({ 'income': [data["x"].sample(20).mean() for _ in range(1000)], 'type': "Mean of 20" }) samples = pd.concat([sample_data, sample_mean_5, sample_mean_20]) samples

Out[4]:

	income	type
18869	70000.00	Data
34424	103000.00	Data
11601	51985.00	Data
39244	46000.00	Data
41527	36000.00	Data
...	...	...
995	58185.00	Mean of 20
996	58664.95	Mean of 20
997	66996.60	Mean of 20
998	65541.50	Mean of 20
999	66092.50	Mean of 20

3000 rows × 2 columns

In [5]:

import matplotlib.pyplot as plt

g = sns.FacetGrid(samples, col="type")
g.map(plt.hist, 'income', bins=40)

import matplotlib.pyplot as plt g = sns.FacetGrid(samples, col="type") g.map(plt.hist, 'income', bins=40)

Out[5]:

<seaborn.axisgrid.FacetGrid at 0x7f05dc630590>