Forecasting of the U.S. Steel Prices with LVAR and VEC Models

Base metal prices, especially steel, play a significant role in industrial economics, making them worth knowing about future values. In most cases, we expect superior performance from multivariate forecasting models comparing univariate methods due to the involvement of explanatory variables system. Standard vector auto regressive model can only capture short-run dynamics because differencing process for non-stationary series that eliminates possible long-run relationship. Instead, performing on levels through auto-regressive framework does not suffers such loss. Moreover, error correction define both short-term and explicitly. These yield more robust forecasts mid-term long-term by investigating relationships simultaneously. The current study aims perform an out-of-sample forecast United States steel prices index 18 months ahead using cointegrated variables. results suggest outperforms regarding mean absolute percentage root square as accuracy measures.