MLOC_LOC Train the local models of a SOM CALLS [mloc] = mloc_loc(y,ny,u,nu,p,mloc) [mloc] = mloc_loc(y,ny,u,nu,p,mloc,usar_bias,sigma) INPUTS y: Output data matrix (Ry x Q) ny: vector of delays for the AR part in the model u: Matrix of input data (Ru x Q) nu: vector of delays in the MA part of the model p: exogenous variables of the index vector (Rp x Q) mloc: SOM structure that models the space generated by selector variables usar_bias: use a bias term in the estimation (default, 1) sigma: width of local neighborhood around a SOM unit used to build the models OUTPUTS mloc: Estructura de SOM con modelos locales SEE ALSO MLOC_SOM, MLOC_SIM, MLOC_PLANOS, MLOC_COMPARA