mgwr.gwr.GWR

GWR.fit(ini_params=None, tol=1e-05, max_iter=20, solve='iwls', searching=False)[source]

Method that fits a model with a particular estimation routine.

Parameters:
ini_betas : array, optional

k*1, initial coefficient values, including constant. Default is None, which calculates initial values during estimation.

tol: float, optional

Tolerence for estimation convergence. Default is 1.0e-5.

max_iter : integer, optional

Maximum number of iterations if convergence not achieved. Default is 20.

solve : string, optional

Technique to solve MLE equations. Default is ‘iwls’, meaning iteratively ( re)weighted least squares.

searching : bool, optional

Whether to estimate a lightweight GWR that computes the minimum diagnostics needed for bandwidth selection (could speed up bandwidth selection for GWR) or to estimate a full GWR. Default is False.

Returns:
:

If searching=True, return a GWRResult instance; otherwise, return a GWRResultLite instance.