Abstract
Vector autoregressive model is a very popular tool in multiple time series analysis. Its parameters are usually estimated by the least squares procedure which is very sensitive to the presence of errors in data, e.g. outliers.
If outliers were present, the estimation results would become unreliable. Therefore in the presented paper we will propose a new procedure for estimating multivariate regression model.
This method is a multivariate generalization of the univariate Least Weighted Squares (LWS) of residuals introduced in [14]. Therefore we will call this estimate Multivariate Least Weighted Squares (MLWS) and we will use it for estimating the coeficients of vector autoregressive model.
We will also perform a simulation study to compare our estimate with LS and robust Multivariate Least Trimmed Squares (MLTS).