bayespecon.logdet.mc

bayespecon.logdet.mc(order, iter, W, rmin=1e-05, rmax=1.0, grid=0.01, random_state=None)[source]

Compute Monte Carlo log-determinant approximation (Barry and Kelley Pace [1999]).

Parameters:
order : int

Number of moments in the stochastic trace expansion.

iter : int

Number of Monte Carlo probes.

W : array-like

Spatial weights matrix.

rmin : float, default=1e-5

Lower bound of the rho grid.

rmax : float, default=1.0

Upper bound of the rho grid.

grid : float, default=0.01

Grid step size.

random_state : int, optional

Seed for reproducibility.

Returns:

Dictionary with rho, lndet, up95, and lo95 vectors.

Return type:

dict