python - statsmodels examples seem not to work -
import statsmodels.formula.api sm import numpy np import pandas url = "http://vincentarelbundock.github.com/rdatasets/csv/histdata/guerry.csv" df = pandas.read_csv(url) df = df[['lottery', 'literacy', 'wealth', 'region']].dropna() print df.head() mod = sm.ols(formula='lottery ~ literacy + wealth + region', data=df) res = mod.fit() print res.summary()
spits error after printing table.
--------------------------------------------------------------------------- typeerror traceback (most recent call last) <ipython-input-4-f69caff21ed0> in <module>() 6 df = df[['lottery', 'literacy', 'wealth', 'region']].dropna() 7 print df.head() ----> 8 mod = sm.ols(formula='lottery ~ literacy + wealth + region', data=df) 9 res = mod.fit() 10 print res.summary() typeerror: from_formula() takes @ least 3 arguments (2 given)
this not seem acceptable behavior. doing wrong?
(the guess in comment wrong)
your version of statsmodels old. documentation , example correct released version of statsmodels 0.5.
the data
keyword has been renamed df
since 0.5.0.dev-1bbd4ca.
so either upgrade, highly recommend, or use old keyword name
mod = sm.ols(formula='lottery ~ literacy + wealth + region', df=df)
should work version have.
Comments
Post a Comment