IdealG¶

class HJCFIT.likelihood.IdealG(*args)[source]¶

Ideal Likelihood.

This object can be instantiated one of several way:

With a QMatrix

>>> matrix = QMatrix(array([...]), 2)
>>> idealg = IdealG(matrix)

af(self, t) → HJCFIT::t_rmatrixAF transitions with respect to time[source]¶

Implements the ideal likelihood:

\[e^{t\mathcal{Q}_{FF}}\mathcal{Q}_{FA}.\]

Parameters:: t – A scalar or something to a numpy array. In the latter case, the return is an array of matrices.

fa(self, t) → HJCFIT::t_rmatrixFA transitions with respect to time[source]¶

Implements the ideal likelihood:

\[e^{t\mathcal{Q}_{AA}}\mathcal{Q}_{AF}.\]

Parameters:: t – A scalar or something to a numpy array. In the latter case, the return is an array of matrices.

property final_vectors¶

Equilibrium vectors for final states.

Computes the right eigenvector of \(\mathcal{G}_{FA}\mathcal{G}_{AF}\), where \(\mathcal{G}_{FA}\) is the laplacian for \(s=0\) of the likelihood.

property initial_vectors¶

Equilibrium vectors for initial states.

Computes the left eigenvector of \(\mathcal{G}_{AF}\mathcal{G}_{FA}\), where \(\mathcal{G}_{AF}\) is the laplacian for \(s=0\) of the likelihood.

laplace_af(self, s) → HJCFIT::t_rmatrixAF transitions with respect to scale[source]¶

Implements the laplace transform of the likelihood:

\[(sI - \mathcal{Q}_{AA})^{-1}\mathcal{Q}_{AF}.\]

Parameters:: s – A scalar or something to a numpy array. In the latter case, the return is an array of matrices.

laplace_fa(self, s) → HJCFIT::t_rmatrixFA transitions with respect to scale[source]¶

Implements the laplace transform of the likelihood:

\[(sI - \mathcal{Q}_{FF})^{-1}\mathcal{Q}_{FA}.\]

Parameters:: s – A scalar or something to a numpy array. In the latter case, the return is an array of matrices.