EOFspaceLeftPatterns ( self )

Returns the left canonical patterns in EOF coordinates

EOFspaceRightPatterns ( self )

Returns the right canonical patterns in EOF coordinates

MCTest (
        self,
        subsamples,
        length,
        )

Monte Carlo test for the temporal stability of the canonical patterns.

Parameters:

subsamples

Number of Monte Carlo subsamples to take

lenght

Length of each subsample (obviously less than the total number od time records)

Returns a tuple with the left and right NumPy arrays containing in each row the congruence coefficient of each subsample obtained patterns with those obtained for the whole dataset.

MCTestCorrelation ( self,  samples )

Monte Carlo test for the significance of the canonical correlations.

The input data are randomly temporally disordered and CCA is performed obtaining a distribution of canonical correlations due to non actually correlated fields but (inherently with the same probability distribution as the original ones)

Parameters:

samples

Number of Monte Carlo subsamples to take

Returns a Numpy array which columns are the different canonical correlations for each MC run.

__init__ (
        self,
        leftfield,
        rightfield,
        retainedeofs=None,
        )

BPCCA constructor.

Parameters:

leftfield

One of the fields (say left) entering the CCA computation (Numpy array). Time is assumed to be its first dimension. The rest are kept as they are.

rightfield

The other field (say right). The first dimension must be the same size as that in leftfield.

retainedeofs

Tuple with two elements containing the number of EOFs (see module svdeofs) to retain in the left and right fields. If not provided, those retaining 70% of the total field variance are selected.

Atributes:

The BPCCA object provides the following accesible atributes. They are only suposed to be accesible for obtaining aditional info, altering their values could be dangerous:

sPCA

Is s SVDEOFs object with the PCA analysis of the left field.

zPCA

Idem for right field.

correlation ( self )

Returns the canonical correlation values

leftAdjointPatterns ( self )

Returns (along the last dimension) the left adjoint canonical patterns

leftExpCoeffs ( self )

Returns the left temporal expansion coefficients

leftPatterns ( self )

Returns (along the last dimension) the left canonical patterns

reconstructedFields ( self,  nccps )

Reconstructs the original fields with the desired number (nccps) of canonical patterns

rightAdjointPatterns ( self )

Returns (along the last dimension) the right adjoint canonical patterns

rightExpCoeffs ( self )

Returns the right temporal expansion coefficients

rightPatterns ( self )

Returns (along the last dimension) the right canonical patterns

varianceFractions ( self )

Returns a tuple with the left and right patterns explained variance fraction