The Relative Performance of VAR and VECM Model

Abstract

Two models are examined in this study: Vector Autoregressive Model (VAR) and Vector Error Correction Model (VECM). Based on three indices: S&P 500, Nikkei 225 (NIKKEI), and Morgan Stanley EAFE (MSCIEAFE or MSCI EAFE) Index, we implement VAR and VECM models, including the pre-estimation diagnostics, model estimation and interpretation and post-estimation tests, etc. By testing, we find that while the VECM model consistently outperforms the VAR model within sample, it is hard to tell which works better out-of-sample. Our examination suggests four conclusions: First, both models do the best job with MSCI EAFE. Second, VECM model does a better job within sample. Third, both models create less error out-of-sample than within sample. Fourth, VECM model does a better job than VAR in S&P 500 and NIKKEI.