statsmodels.tsa.vector_ar.vecm.VECMResults.orth_ma_rep¶

VECMResults.orth_ma_rep(maxn=10, P=None)[source]¶

Compute orthogonalized MA coefficient matrices.

For this purpose a matrix P is used which fulfills ${\mathrm{\Sigma }}_u=P{P}^{\prime }$. P defaults to the Cholesky decomposition of ${\mathrm{\Sigma }}_u$

Parameters:¶

maxnint

Number of coefficient matrices to compute

Pndarray (neqs x neqs), optional

Matrix such that ${\mathrm{\Sigma }}_u=P{P}^{\prime }$. Defaults to Cholesky decomposition.

Returns:¶

coefsndarray (maxn x neqs x neqs)