Pandas: OLS regression does not output an intercept

Question

I have the following ouput from a Pandas pooled OLS regression. The only problem is that I'm not sure where the intercept is. In a regression there is always an intercept that is usually listed before the exogenous variables, ie Y = a + ßx1 + ßx2 + error_term I do not see it in my regression. I used the suggestion from ayhan X = add_constant(X) but somehow I get the feeling that I am messing something up (in an obvious way) with the syntax. I know that this is not rocket science. Can someone tell me what I am missing?

import numpy as np
import pandas as pd
from pandas import DataFrame, Series
import statsmodels.formula.api as sm
from sklearn.linear_model import LinearRegression
import scipy, scipy.stats
import matplotlib.pyplot as plt
import matplotlib
matplotlib.style.use('ggplot')
from statsmodels.api import add_constant



 X = add_constant(X)
Y = df['billsum_support']

X = df[['direct_expenditures','indirect_expenditures', 'years_exp', 'leg_totalbills',\
    'log_diff_rgdp', 'unemployment',  'expendituresfor']]

result = sm.OLS( Y, X ).fit()
result.summary()

 OLS Regression Results Dep. Variable:  billsum_support     R-squared:  0.663
Model:  OLS     Adj. R-squared:     0.663
Method:     Least Squares   F-statistic:    3932.
Date:   Sun, 08 May 2016    Prob (F-statistic):     0.00
Time:   22:38:33    Log-Likelihood:     -12561.
No. Observations:   12008   AIC:    2.513e+04
Df Residuals:   12002   BIC:    2.518e+04
Df Model:   6       
Covariance Type:    nonrobust       
    coef    std err     t   P>|t|   [95.0% Conf. Int.]
direct_expenditures     4.575e-05   4.02e-06    11.377  0.000   3.79e-05 5.36e-05
indirect_expenditures   -2.147e-05  6.93e-06    -3.099  0.002   -3.5e-05 -7.89e-06
years_exp   0.0030  0.001   5.595   0.000   0.002 0.004
leg_totalbills  0.0052  0.000   11.160  0.000   0.004 0.006
log_diff_rgdp   1.0325  0.178   5.805   0.000   0.684 1.381
unemployment    0.1052  0.001   70.744  0.000   0.102 0.108
expendituresfor     2.428e-05   3.57e-06    6.797   0.000   1.73e-05 3.13e-05
Omnibus:    2994.033    Durbin-Watson:  0.837
Prob(Omnibus):  0.000   Jarque-Bera (JB):   19159.354
Skew:   1.042   Prob(JB):   0.00
Kurtosis:   8.827   Cond. No.   1.54e+16

Answer 1

You need to explicitly tell statsmodels to fit an intercept. Update your independent variables with statsmodels.api.add_constant :

from statsmodels.api import add_constant
Y = df['billsum_support']
X = df[['direct_expenditures','indirect_expenditures', 'years_exp', 'leg_totalbills',\
    'log_diff_rgdp', 'unemployment',  'expendituresfor']]
X = add_constant(X)
result = sm.OLS( Y, X ).fit()