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_eq
int
number of equations
- WS
spatial
weights
matrix
in
sparse
form
- bigy
dictionary
with y values
- bigX
dictionary
with X values
- bigE
array
n x n_eq matrix of residuals by equation
- bigYP
array
n x n_eq matrix of predicted values by equation
- sig
array
cross-equation error covariance matrix
- varb
array
variance-covariance matrix for b coefficients (inverse of Ibb)
- n_eq
- Returns:
- (LMlag,n_eq,pvalue)
tuple
with value of statistic (LMlag), degrees of freedom (n_eq) and p-value
- (LMlag,n_eq,pvalue)