Correlations between commodities

Identifying what one market’s move means to another market is important. At DePutter Publishing we explore these connections with various tools. Things can get complex, and visualizing data and their relations is helpful.

Below you can see a plot of the correlation parameters calculated between the near monthly futures of a number of commodities. The period considered goes from January 2006 to January 2020.

Correlation matrix between the futures of a number of commodities. Each square contains a subplot of the joint evolution of the futures of two commodities. A .pdf, high-resolution version, which can be easily zoomed in, is available here. Guidance in interpreting the figure is provided in the main text. Copyright DePutter Publishing 2020.

The square at the intersection of a each row and column contains a joint plot of the two futures of the two corresponding commodities, meant to emphasize correlated behaviour, when present (a high-resolution version of the figure is available here; use a .pdf reader to zoom in). The correlation value appears in the upper left corner of each square. Such a value is reflected also in the color of the square, as we now proceed to explain.

The correlation value varies between -1 and +1. The closer to +1 the correlation value (correspondingly, the closer to green the background of the square), the more correlated the (futures of the) two commodities, that is, they have a tendency to go up or down at the same time. The closer to -1 the correlation value (correspondingly, the closer to purple the background of the square), the more anticorrelated the (futures of the) two commodities, that is, if one goes up, the other has a tendency to go down, and viceversa. The closer to 0 the correlation value (correspondingly, the closer to white the background of the square), the less related the (futures of the) two commodities are, that is, they tend to go up and down relatively independently. The blocks on the diagonal plot each commodity against itself, so that all the points lie on a straight line and the correlation is perfect and equal to 1.

Obviously, a single number cannot capture the nuances of the relations between two commodities, let alone the influence of the price of one commoditiy on the whole of field of agriculture (see our previous post for an example of more in-depth analysis). Nonetheless, it is useful to use such a number as an indicator, and it is also useful to be able to do so at glance, e.g., using correlation plots like the above.

Our services like the Market Advisory Service newletter and the Ag-Alert newsletter provide a timely and in-depth analysis of single commodity markets and of the interplay between such commodities. For a free trial, please use the links below.

Try a FREE 3 week trial

Ag-Alert

Ag-Alert


Like a full-time professional marketing consultant for your farm.

Try a FREE 3 week trial

The DePutter Market Advisory Service

The DePutter Market Advisory Service


Make smart decisions for your farm.