The Bootstrap¶

In [1]:

from sklearn.utils import resample
import pandas as pd

from sklearn.utils import resample import pandas as pd

In [2]:

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

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

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]:

results = []
for _ in range(1000):
    sample = resample(loans_income)
    results.append(sample["x"].mean())

res_df = pd.Series(results)
res_df

results = [] for _ in range(1000): sample = resample(loans_income) results.append(sample["x"].mean()) res_df = pd.Series(results) res_df

Out[3]:

0      68766.81156
1      68770.10790
2      68848.16744
3      68946.84508
4      68710.84424
          ...     
995    68722.73528
996    68428.32996
997    68539.88930
998    68657.13478
999    68880.52984
Length: 1000, dtype: float64

In [4]:

print(f"Original median: {loans_income["x"].mean()}")
print(f"Bootstrap method median {res_df.mean()}")
print(f"Bias: {res_df.mean()- loans_income["x"].mean()}")
print(f"Std. Error: {res_df.std()}")

print(f"Original median: {loans_income["x"].mean()}") print(f"Bootstrap method median {res_df.mean()}") print(f"Bias: {res_df.mean()- loans_income["x"].mean()}") print(f"Std. Error: {res_df.std()}")

Original median: 68760.51844
Bootstrap method median 68761.05207012
Bias: 0.5336301200004527
Std. Error: 142.41785215344774