Recently food commodity prices have skyrocketed due to severe drought in the U.S. The price of wheat rose 45% from June to August.
Could we have predicted and profited from this price increase?
The previous article introduced the basics of the IKnowFirst stock forecast algorithm. The charts below (Figs. 1 and 2.) present the system actual performance on wheat price prediction. The thick red line shows the actual Dow Jones-UBS Wheat index ^DJUBWHTR. This is one of the possible vehicles to play commodities.
The thin blue line on the chart is a signal line. The positive or negative (Up or Down) signals of the forecast were added to the actual last known price at the time of forecast. Thus, when the signal line is above the actual line, it means BUY, if below, then SELL. Each point on this chart was taken from the actual daily forecast published in the morning before the next market open. The seven days forecast on this chart is one of the six times range forecasts that the system outputs, from three days to one year ahead.
Fig 1:(click to enlarge)
Results and discussion
One can see from the Fig. 1 that the first significant up signal occurred at the end of May 2012, see arrow. As Fig. 2 shows, the 7 days signal was confirmed by other time ranges signals. The signals kept strongly up all through June, while the wheat prices were still low. That would be the ideal entry time. Getting in after that is much more risky, even though the signal was still up.
These charts could form a basis of a simple strategy. The rules are: Buy when the combined signal line of all time ranges is above the actual line, sell when below. In this case this strategy would be profitable.
It's important to pay attention to the signal strength. It's preferable to enter the market on strong signals. The October 2011 down signals and May-June 2012 up signals were strong (Fig. 2). This is opposed to the weak January 2012 down signals. Had one followed up on them, he would be in the loss in the short term, and would come back even after three months. Another reason not to enter at that point: the price of wheat was already strongly down by a lot, and further shorting would bring a risk of rebound.
More complicated strategies can also be devised, such as linear combinations of different time range signals, signal strength filters, and signals in relation to predictability filters. For example, Fig. 2 shows not the straight signal, but the signal multiplied by predictability. This allows enhancing signals with strong predictability, and suppressing ones with weak predictability.
Such strategies can reduce the risk, but not eliminate it entirely. Risk management is of course essential, but this is a different subject entirely.
How to invest (speculate) in food commodities?
The most common method is through futures and options contracts. One should be aware of the contango, backwardation, the specific risks and costs due to the fixed term of contracts. A less common method is to invest in the commodity based ETFs. There are many such ETFs and ETN's specializing in agriculture, wheat and other grains, corn, and soybeans (see list below).
Agricultural Commodity ETFs short list
Wheat (NYSEARCA:JJG), Corn (^DJUBCNTR), soy oil (^DJUBBOTR), coffee (NYSEARCA:JO), cocoa (NYSEARCA:NIB), sugar (^DJUBSSB), cotton, (^DJUBCTTR). PowerShares DB Commodity Index Tracking (NYSEARCA:DBC), PowerShares DB Agriculture (NYSEARCA:DBA), also ^DJUBAGTR, CORN, MOO, COW. General commodity indexes such as ^DJUBS, Basic Materials (^DJUSBM). The IKnowFirst system keeps track of most of these food commodities and also of energy (OIL), metals (GLD, SLV) and more.
Wheat price forecast: What is the next trend?
Regardless, the fact is that food commodities are strongly cyclical. Soon there might be opportunities in going short. The current 90 days predictability of wheat of IKnowFirst system is 0.67. What is predictability and how it relates to the signal is explained in the previous article.
Disclosure: I have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.