Descriptive Stats
import numpy as np
import math
import pandas as pd
x=[x1,x2,x3,...]
x_df=pd.DataFrame({'column_name':x})| procedure | numpy | pandas |
|---|---|---|
| frequency counts |
np.unique(x,return_counts=True)
|
x_df.value_counts
|
| size |
np.size(x)
|
x_df.count
|
| min |
np.amin(x)
|
x_df.min
|
| max |
np.amax(x)
|
x_df.max
|
| mean |
np.mean(x)
|
x_df.mean
|
| mode | NA |
x_df.mode
|
| median |
np.median(x)
|
x_df.median
|
| sample variance |
np.var(x,ddof=1)
|
x_df.var
|
| pop variance |
np.var(x)
|
x_df.var(ddof=0)
|
| sample std |
np.std(x,ddof=1)
|
x_df.std
|
| pop std |
np.std(x)
|
x_df.std(ddof=0)
|
| weighted mean |
np.average(x,weights=freq)
| NA |
| weighted sample std |
(np.cov(x,fweights=freq,ddof=1))**(1/2)
| NA |
| weighted pop std |
(np.cov(x,fweights=freq))**(1/2)
| NA |
| percentiles |
np.quantile(x,p)
|
x_df.quantile(p)
|
| summary stats | NA |
x_df.describe
|
| factorial, n! |
math.factorial(n)
| NA |
| n random integers from a to b |
np.random.randint(a,b+1,size=n)
| NA |
| weighted mean |
np.average(x,weights=y)
| NA |
| Weighted std |
(np.cov(x,aweights=y,ddof=0))**(1/2)
| NA |
| correlation coef. |
np.corrcoef(x,y)
|
x_df.corr
|
NA = Not Available
Summary statistics by grouping with categorical columns:
x_df.groupby(['level','gender']).agg({'test1_score': ['mean','std','size']})
Adapted from Python for Introductory Statistics, by Simon Aman (Truman College, City Colleges of Chicago), licensed under CC BY 4.0. Changes were made: reformatted as an accessible XYZ web edition with live in-browser code cells. License: CC-BY-4.0.