qspec.analyze.covariance_matrix  (  cov = None ,  sigma = None ,  corr = None ,  k = None ,  n = None  )[source]

Helper function to construct covariance matrices $\mathbf{\Sigma}_i\in\mathbb{R}^{n\times n}$ from standard deviations $\vec{\sigma}_i\in\mathbb{R}^n$ and correlation matrices $\mathbf{\rho}_i\in\mathbb{R}^{n\times n}$.

Parameters:

covarray_like | None

The covariance matrices $\mathbf{\Sigma}_i$. Must have shape (k, n, n), where k is the number of data points and n the number of dimensions of each data point. If None, the other parameters are used. Else, all other parameters are ignored.

sigmaarray_like | None

The standard deviations of the data vectors. Must have shape (k, n). If None, all diagonal elements of the covariance matrix equal 1.

corrarray_like | None

The correlation matrices of the data vectors. Must have shape (k, n, n). If n == 2 it can have shape (k, ). If None, all off-diagonal elements of the covariance matrix are 0.

kint | None

The number of data points. It is omitted if sigma or corr is specified.

nint | None

The number of dimensions $n$ of each data point. It is omitted if sigma or corr is specified.

Returns:

covndarray

Covariance matrices $\mathbf{\Sigma}_i$ constructed from sigma and corr.