ABSTRACT
The vector autoregressive (VAR) approach may help supplement of the previous approach in forecasting the prices, by directly providing forecast values of steel prices. The use of vector autoregression methods may overcome some of the problems, but some of the difficulties will remain. A VAR model consists of a system of equations in which each variable is regressed on its own lagged values and on the lagged values of all the other variables in the model. The Bayesian approach combines a set of prior beliefs with the traditional VAR methods. In general, a VAR model expresses the current values of the endogenous variables solely as a function of an intercept variable and lagged values of the endogenous variables. The forecast of the five steel products until 2000 was conducted using a Bayesian VAR model. Unexpected economic boom or recession in some countries also affect resource prices and exchange rates, which in turn changes steel prices.
