Package mvpa :: Package measures :: Module glm :: Class GLM
[hide private]
[frames] | no frames]

Class GLM

source code


General linear model (GLM).

Regressors can be defined in a design matrix and a linear fit of the data is computed univariately (i.e. indepently for each feature). This measure can report 'raw' parameter estimates (i.e. beta weights) of the linear model, as well as standardized parameters (z-stat) using an ordinary least squares (aka fixed-effects) approach to estimate the parameter estimate.

The measure is reported in a (nfeatures x nregressors)-shaped array.

Nested Classes [hide private]

Inherited from misc.state.ClassWithCollections: __metaclass__

Instance Methods [hide private]
 
__init__(self, design, voi='pe', **kwargs)
Initialize
source code
 
_call(self, dataset)
Computes a per-feature-measure on a given Dataset.
source code

Inherited from base.FeaturewiseDatasetMeasure: __repr__, combiner

Inherited from base.FeaturewiseDatasetMeasure (private): _postcall

Inherited from base.DatasetMeasure: __call__, null_dist, transformer, untrain

Inherited from misc.state.ClassWithCollections: __getattribute__, __new__, __setattr__, __str__, reset

Inherited from object: __delattr__, __format__, __hash__, __reduce__, __reduce_ex__, __sizeof__, __subclasshook__

Class Variables [hide private]
  pe = StateVariable(enabled= False, doc= "Parameter estimates (...
  zstat = StateVariable(enabled= False, doc= "Standardized param...

Inherited from base.FeaturewiseDatasetMeasure: base_sensitivities

Inherited from base.DatasetMeasure: __doc__, null_prob, null_t, raw_results

Inherited from misc.state.ClassWithCollections: _DEV__doc__, descr

Properties [hide private]

Inherited from object: __class__

Method Details [hide private]

__init__(self, design, voi='pe', **kwargs)
(Constructor)

source code 
Initialize
Parameters:
  • design, array(nsamples, x, nregressors) - GLM design matrix.
  • voi, 'pe', |, 'zstat' - Variable of interest that should be reported as feature-wise measure. 'beta' are the parameter estimates and 'zstat' returns standardized parameter estimates.
Overrides: object.__init__

_call(self, dataset)

source code 

Computes a per-feature-measure on a given Dataset.

Behaves like a DatasetMeasure, but computes and returns a 1d ndarray with one value per feature.

Overrides: base.DatasetMeasure._call
(inherited documentation)

Class Variable Details [hide private]

pe

Value:
StateVariable(enabled= False, doc= "Parameter estimates (nfeatures x n\
parameters).")

zstat

Value:
StateVariable(enabled= False, doc= "Standardized parameter estimates (\
nfeatures x nparameters).")