def norm_resid(self):
"""
Residuals, normalized to have unit length and unit variance.
Returns
-------
An array wresid/sqrt(scale)
Notes
-----
This method is untested
"""
if not hasattr(self, 'resid'):
raise ValueError, 'need normalized residuals to estimate standard\
deviation'
return self.wresid * recipr(np.sqrt(self.scale))
def t_test(self, r_matrix, scale=None):
"""
Compute a tcontrast/t-test for a row vector array.
Parameters
----------
r_matrix : array-like
A length p row vector specifying the linear restrictions.
scale : float, optional
An optional `scale` to use. Default is the scale specified
by the model fit.
scale : scalar
Examples
--------
>>> import numpy as np
>>> import scikits.statsmodels as sm
>>> data = sm.datasets.longley.load()
>>> data.exog = sm.add_constant(data.exog)
>>> results = sm.OLS(data.endog, data.exog).fit()
>>> r = np.zeros_like(results.params)
>>> r[4:6] = [1,-1]
>>> print r
[ 0. 0. 0. 0. 1. -1. 0.]
r tests that the coefficients on the 5th and 6th independent
variable are the same.
>>>T_Test = results.t_test(r)
>>>print T_test
<T contrast: effect=-1829.2025687192481, sd=455.39079425193762, t=-4.0167754636411717, p=0.0015163772380899498, df_denom=9>
>>> T_test.effect
-1829.2025687192481
>>> T_test.sd
455.39079425193762
>>> T_test.t
-4.0167754636411717
>>> T_test.p
0.0015163772380899498
See also
---------
t : method to get simpler t values
f_test : for f tests
"""
r_matrix = np.squeeze(np.asarray(r_matrix))
if self.normalized_cov_params is None:
raise ValueError, 'Need covariance of parameters for computing \
T statistics'
if r_matrix.ndim == 1:
if r_matrix.shape[0] != self.params.shape[0]:
raise ValueError, 'r_matrix and params are not aligned'
elif r_matrix.ndim >1:
if r_matrix.shape[1] != self.params.shape[0]:
raise ValueError, 'r_matrix and params are not aligned'
_t = _sd = None
_effect = np.dot(r_matrix, self.params)
_sd = np.sqrt(self.cov_params(r_matrix=r_matrix))
if _sd.ndim > 1:
_sd = np.diag(_sd)
_t = _effect * recipr(_sd)
return ContrastResults(effect=_effect, t=_t, sd=_sd,
df_denom=self.model.df_resid)