spreg.surLMlag

spreg.surLMlag(n_eq, WS, bigy, bigX, bigE, bigYP, sig, varb)[source]

Lagrange Multiplier test on lag spatial autocorrelation in SUR

Parameters:
n_eqint

number of equations

WSspatial weights matrix in sparse form
bigydictionary

with y values

bigXdictionary

with X values

bigEarray

n x n_eq matrix of residuals by equation

bigYParray

n x n_eq matrix of predicted values by equation

sigarray

cross-equation error covariance matrix

varbarray

variance-covariance matrix for b coefficients (inverse of Ibb)

Returns:
(LMlag,n_eq,pvalue)tuple

with value of statistic (LMlag), degrees of freedom (n_eq) and p-value