Objective research and reviews to aid investing decisions
 Looking for information on debt consolidation loans? Want to know more about reverse mortgages? |
After reviewing our blog entry of 1/29/08 that finds no evidence supporting Charles Nenner's belief in a relationship between sunspots and the stock market, reader Henry Bee of Vancouver, Canada asks:
"Sunspot activity does have a direct relationship to weather. Could one speculate on the natural gas market or the agriculture market using the sunspot cycles?"
Using monthly sunspot counts and monthly U.S. wheat prices for January 1950 through November 2007 (695 months) and monthly U.S. wellhead natural gas prices for January 1976 through November 2007 (383 months), we find that...
The National Weather Service notes:
"...[T]he jury is still out on how much sunspots can (or do) affect the Earth's climate. Times of maximum sunspot activity are associated with a very slight increase in the energy output from the sun. Ultraviolet radiation increases dramatically during high sunspot activity, which can have a large effect on the Earth's atmosphere. From the mid 1600s to early 1700s, a period of very low sunspot activity (known as the Maunder Minimum) coincided with a number of long winters and severe cold temperatures in Western Europe, called the Little Ice Age. It is not known whether the two phenomena are linked or if it was just coincidence. The reason it is hard to relate maximum and minimum solar activity (sunspots) to the Earth's climate, is due to the complexity of the Earth's climate itself."
If there were relationships, we hypothesize that a high sunspot count (warm weather) would depress natural gas price (lower heating demand) and depress wheat price (support good crops).
The following chart compares the monthly sunspot count to the contemporaneous monthly U.S. wellhead natural gas price (log scale) during January 1976 through November 2007. Visual inspection reveals no consistent relationship between the two series. It does not appear that natural gas price has valleys (peaks) at sunspot peaks (valleys).
For a more precise measurement, we investigate whether changes in sunspot activity relate to changes in the price of natural gas.
The following scatter plot relates the change in U.S. wellhead natural gas price over three-month periods to change in sunspot count over the same periods. We have winnowed the sample such that the three-month periods do not overlap (127 surviving sample periods). Visual inspection suggests no relationship. The Pearson correlation is 0.03 and the R-squared statistic is 0.00, confirming no relationship.
When we offset such that changes in sunspot activity lead changes in natural gas price by three months, the Pearson correlation is -0.08 and the R-squared statistic 0.01. There is no significant leading relationship.
What about a longer-term relationship?
The next scatter plot relates the change in U.S. wellhead natural gas price over 12-month periods to change in sunspot count over the same periods. We have further winnowed the sample such that the 12-month periods do not overlap (only 31 surviving sample periods). Visual inspection suggests no relationship. The Pearson correlation is -0.06 and the R-squared statistic is 0.00, confirming no relationship. However, sample size is very small.
When we offset such that changes in sunspot activity lead changes in natural gas price by 12 months, the Pearson correlation is 0.09 and the R-squared statistic 0.01. There is no significant leading relationship.
How about sunspots and wheat?
The following chart compares the monthly sunspot count to the contemporaneous monthly U.S. wheat price (log scale) during January 1950 through November 2007. Visual inspection reveals no consistent relationship between the two series. There are some cases in which wheat prices are high (low) when sunspot activity is low (high), but there are also contradictory data.
For a more precise measurement, we investigate whether changes in sunspot activity relate to changes in the price of wheat.
The following scatter plot relates the change in U.S. wheat price over three-month periods to change in sunspot count over the same periods. We have winnowed the sample such that the three-month periods do not overlap (231 surviving sample periods). Visual inspection suggests no relationship. The Pearson correlation is -0.04 and the R-squared statistic is 0.00, confirming no relationship.
When we offset such that changes in sunspot activity lead changes in wheat price by three months, the Pearson correlation is -0.02 and the R-squared statistic 0.00. There is no leading relationship.
What about a longer-term relationship?
The final scatter plot relates the change in U.S. wheat price over 12-month periods to change in sunspot count over the same periods. We have further winnowed the sample such that the 12-month periods do not overlap (only 57 surviving sample periods). Visual inspection suggests no relationship. The Pearson correlation is -0.06 and the R-squared statistic is 0.00, confirming no relationship. However, sample size is small.
When we offset such that changes in sunspot activity lead changes in wheat price by 12 months, the Pearson correlation is -0.17 and the R-squared statistic 0.03. These results suggest a weak negative relationship, with sunspot variation explaining 3% of the variation in wheat price one year ahead. However, a few contradictory data points added to such a small sample might substantially change the statistics.
In summary, evidence does not support using sunspot activity to speculate on natural gas and wheat prices over periods of a quarter or a year.
For related research, see Blog Synthesis: Calendar Effects.
