Multivariate Breusch Godfrey Lagrange Multiplier tests in Python

问题

I understand that in the package statsmodel has many statistical functions that enable one to test for many issues including Breusch Godfrey Lagrange test as described here

However, as far as I am concerned this only do the job for univariate case and not the multivariate case.

For example, consider I have a 2 diminsional dataset say data

from statsmodels.tsa.api import VAR
import statsmodels.api as sm,statsmodels as sm1

data= np.random.random((108, 2))
Model=VAR(data)
results = Model.fit(1)
sm.stats.diagnostic.acorr_breusch_godfrey(results)

Output: 
in acorr_breusch_godfrey(results, nlags, store)
    501         nlags = int(nlags)
    502 
--> 503     x = np.concatenate((np.zeros(nlags), x))
    504 
    505     #xdiff = np.diff(x)

ValueError: all the input arrays must have same number of dimensions

Hence, obviously, this only work for one dimensional, is anyone aware how to test for Breusch Godfrey in the multivariate setting?

来源：https://stackoverflow.com/questions/46302347/multivariate-breusch-godfrey-lagrange-multiplier-tests-in-python