DeepVu’s Commodity Forecasting is a suite of ML models that forecast the price of the Futures contract of a commodity that is traded on the Chicago Mercantile Exchange or Shanghai Futures market. We have models for forecasting the price of industrial metals, precious metals, agricultural commodities, and chemicals, etc. DeepVu’s models are trained on a historical data-set of a minimum of 10years of daily data, and we typically forecast with weekly cadence or monthly cadence. The forecast horizon varies by commodity but we often deliver a forecast ranging from 12 weeks out to 52 weeks out. 

The farther out the forecast horizon, the higher the price of the model.
DeepVu’s models take into account a lot of external signals beyond the price action of the instrument being forecasted.
We refresh and tune our models either once every 3 weeks or once every 6 weeks depending on the pricing plan.

This model is for Aluminum Shanghai Futures, for 1 year out with weekly cadence.