Forecasting cotton prices using statistical methods has advantages and disadvantages. The limitations are mainly a matter of data: for structural models, we often don’t have data to model the variables that are really influencing prices. Even if we do have data, our model prediction of price is tied up with the historical set of data that we used to generate the prediction. This can break down when we are in exceptional time periods such as 2010-11.
The advantages of using regression forecasts is that it is a generally acceptable, repeatable, systematic method of measuring the influence of important (or so we hypothesize) variables on cotton prices. For example, we specify that the ICE New York most active cotton futures price is a function of 1) the monthly average net long position (in futures contracts) of hedge funds, 2) U.S. cotton ending stocks (revised monthly by USDA), and 3) a dummy variable accounting for the abnormally high price spike of 2010 and 2011. Using monthly observations between 2000 and formal results of this regression are summarized here:
A quick interpretation of this output is as follows. First, this relationship of the specified independent variables does a fairly decent job explaining most (82%) of the variability in the most active ICE cotton futures contract.
Concerning those variables, the results indicate that U.S. cotton farm price move in the opposite direction of U.S. ending stocks-to-use. More specifically, for every 1% increase in U.S. ending stocks, the most active ICE cotton futures price would be expected to weaken by 0.955 cents per pound. The price spike dummy variable was also significant and positive.
The beta coefficient for the Hedge Fund net position is significant and positive. This estimate means that for a every thousand contract increase in the Hedge Fund net long position, the most active cotton futures contract price will increase by 0.19 cents per pound. This suggests that hedge fund buying and selling can be a strong influence on cotton futures during a major rallies or sell-offs. For example, this regression modeling lets us say specifically that fund buying was associated with about three quarters of the twelve cent rally that occurred between December, 2013 and April, 2014. Likewise, hedge fund selling between July and September was associated with eleven cents worth of the 24 cent decline in cotton futures over that time. These results do not prove or even imply causation. They are meant to draw the attention of grower-hedgers to the potential influence and patterns of fund buying and selling.
From a price forecasting standpoint, this model predicts a near term futures price of 62.79 cents per pound for the most active ICE cotton futures contract. That is within the trading range of Dec’14 all during the fall and through November 10. The 62.79 cent prediction is shown by the black line on the far right side of the chart below. The associated risk bounds imply, unhelpfully, that there is a 95% probability of the average monthly futures price lies within a fifty cent band.
