you can use Autogressive Integrated Moving Average (ARIMA) for forecastig the prices of Copper

There are some global cyclical trends that seem to be predictable and amenable to model building.

Good day Srikanth

Thank you for the response. The concern with using an average or moving average is the reliability when you are attempting to forecast for periods of 5-20 years in the future. Logically would assume some real medium or long term price as an estimate of the average future price?

Regards John

Good day Emmanuel

The observation concerning global commodity cycles is appropriate, but how do you estimate where you are in a cycle right now? If you assume prices have peaked and are about to decline and the inverse happens, the forecast estimate becomes more inaccurate. So it comes back to your second point about an amenable model. Which is the most reliable model to use for forecasting commodity prices? IMF/World Bank use global economic growth, other specialized firms supply and demand forecasts based upon fundamental analysis. Financial institutions, such as research analysts also prepare estimates, would assume based on some fundamental analysis in combination with the futures market. Question is which of these approaches is the most reliable?

Regards John

CME trades copper futures with up to 5 years maturity. It is a "market" view or how traders/hedgers view the future prices of this commodity. Clearly on other side the review of current / new mines production capacity and demand from major industries should determine price equilibrium. Copper is an easy to store to mitigate any surges in demand.

Good day Leon

Thank you for your reply. Definitely concede the point on CME, but it raises some challenges of its own. If a copper producer is of the view that the future copper price will increase, they may be reluctant to hedge the price risk. Numerous shareholders prefer mining organizations not to hedge the future price risk, ensuring the "blue sky" potential. Hedging is not a cost-less exercise (e.g. margin calls) and with the accounting requirements for financial instruments, any adverse price movements impact P&L impacting share prices. The volume of copper traded on the metal exchanges by far exceeds the physical market, so it is perhaps possible to view these more as speculator avenues than reflecting the underlying physical demand-supply market. Most tellingly, any futures price is dependent on the spot price, so any speculative trading can vary the future price substantially.

For the long term demand supply fundamentals, who would you recommend, e.g. Wood Mackenzie?

The more I look into the problem, the more it would seem the answer does not lie in a specific source, but rather taking a consensus view from the forecasts of several different approaches. The tendency of mining organizations to rely predominantly on equity research analysts and Wood Mackenzie appears to be biased?

Regards John

Wood Mackenzie is not alone in this field. S&P Global Platts is another respectful place. Plus research departments in banks dealing with commodities. They publish reports as well. The bottom line is to figure out who are major consumers for copper - telecommunication, construction, ... based on my limited knowledge there is no real replacement (i.e by aluminum) for wires (at least in construction) and now for electric car batteries. There are sources of estimates for future need for the metal, like copper.org. On other hand, there are few major copper producers (like Freeport, Rio-Tinto ,..). In their annual reports they publish their mining capacity and geographic location of mines as well as discovered reserves. I guess this is a starting point. And I guess CME quotes to view consensus. Those are my 2 cents.

Most probably, I guess, a creative statistical modeling via regression analysis, the latest price movements (last 10 years) given.

Good day Leon

All the sources you reference are definitely important, it is just they tend to have a short-term forecast, 3 years or so and then revert to son long-term forecast assumed to be a real price to be escalated by US inflation. Because of the dominance of the big specialist firms the financial institutions tend to revert to these forecasts as the long term benchmark.

What I am leaning towards if I took a weighted consensus from all these sources, would it be a better estimate than any of the individual sources? In taking a consensus forecast the intention would be to look at different methodologies used to prepare the forecasts.

I do not have access to historical forecast from the various sources and have encountered some resistance in obtaining previous forecasts, but ideally would like to do this as a preliminar stage to estimate the optimal weightings.

The point is from a mining perspective, the investment decision is made and is then followed by many years (even decades) of production that was premised on an estimated copper price. The question ultimately is how accountable should the mining organization be for the copper price estimate used?

Regards John

I will suggest you to used "Dow Theory" . It is very effective in predicting price movements for long term.

Second option can be Elliot Wave Theory.

Feel free to contact me if you need any help or collaboration with your research.