Logo

statsmodels.regression.linear_model.OLS

class statsmodels.regression.linear_model.OLS(endog, exog=None)[source]

A simple ordinary least squares model.

Methods

inherited from regression.GLS

Parameters:

endog : array-like

1d vector of response/dependent variable

exog: array-like :

Column ordered (observations in rows) design matrix.

Notes

OLS, as the other models, assumes that the design matrix contains a constant.

Examples

>>> import numpy as np
>>>
>>> import statsmodels.api as sm
>>>
>>> Y = [1,3,4,5,2,3,4]
>>> X = range(1,8) #[:,np.newaxis]
>>> X = sm.add_constant(X)
>>>
>>> model = sm.OLS(Y,X)
>>> results = model.fit()
>>> results.params
array([ 0.25      ,  2.14285714])
>>> results.tvales
array([ 0.98019606,  1.87867287])
>>> print results.t_test([0,1])
<T test: effect=2.1428571428571423, sd=1.1406228159050935, t=1.8786728732554485, p=0.059539737780605395, df_denom=5>
>>> print results.f_test(np.identity(2))
<F test: F=19.460784313725501, p=0.00437250591095, df_denom=5, df_num=2>

Attributes

weights scalar Has an attribute weights = array(1.0) due to inheritance from WLS.
See regression.GLS    

Methods

fit([method]) Full fit of the model.
hessian(params) The Hessian matrix of the model
information(params) Fisher information matrix of model
initialize()
loglike(params) The likelihood function for the clasical OLS model.
predict(params[, exog]) Return linear predicted values from a design matrix.
score(params) Score vector of model.
whiten(Y) OLS model whitener does nothing: returns Y.

Attributes

endog_names
exog_names

Previous topic

Technical Documentation

Next topic

statsmodels.regression.linear_model.OLS.fit

This Page