Methodology

The three drivers that influence the base metal price.

The three drivers that influence base metal prices are:

Metal Stock levels and stock change

Global IP cycles

The strength of the US Dollar

MPA favours three modelling methods:

MPA favours three modelling methods for exploratory and final model preparation.

– LASSO (least absolute shrinkage and selection operator) and Ridge Regression are favoured over multivariate linear regression as these techniques are more effective in reducing model complexity, reduce over-fitting and are also helpful with dealing with colinearity in the drivers.

– Multivariate Adaptive Regression Splines which help to deal with non-linear relationships and pinch points of the individual drivers

– Classification and Regression Trees which help to subset the data into different regions that may exhibit different behaviour (e.g. very low stocks or very high y-o-y IP).

MPA uses these methods to build a factor model of the metal price dependent on the factors (or drivers) of Stock, Stock Change, IP cycles and Dollar strength. The modelling time period is between 2015 to the present.

Using the Factor Model to forecast three years ahead.

Once the Factor Model is built, the forecasts of the three underlying drivers can be input into the model to obtain the metal price forecast.