This page outlines Tutorial 1 [link]
The following are the commands used:
import pandas as pd ver=pd.read_csv("df.csv") print ver.head(3) print len(ver) print ver.columns print ver.dtypes print ver['RegionState'] print ver['Infant MR'] print ver[['RegionState','Infant MR','Heart Disease DR']] print ver.describe() print ver.sort(['Infant MR']).head(3) print ver.sort(['Infant MR'],ascending=False).head(3) print ver.corr() print ver[['Infant MR','Heart Disease DR']].corr() print ver.cov() print ver['Infant MR']> 4.8 print ver[(ver['Infant MR'] > 7.0)] print ver[(ver['Infant MR'] > 7.0) & (ver['Stroke DR'] > 45.0 )] import statsmodels.api as sm print sm.OLS(ver['Infant MR'], ver['Heart Disease DR']).fit().summary() import numpy as np m, b = np.polyfit(ver[‘Infant MR’], ver[‘Heart Disease DR’], 1) print 'Infant MR = '+str(round(m,3))+' * Heart Disease DR + ',str(round(b,3))