I’m Lovin’ It

Deciphering the future often involves looking for activity in the past that is likely to repeat itself. For this reason, statistical correlation sits at the heart of forecasting. There are two dangers to be aware of in this. Firstly, when the future no longer resembles the past, due to a change in the patterns of activity, perhaps driven by technology and changing social mores. Secondly, when the correlation is pure chance, and not due to causation. This week, I read an article by a financial analyst exploring ‘The McRib effect’, a correlation between the availability of the seasonal McRib in McDonald’s and the performance of the S&P 500 (0.07% higher on those days). For those looking for further examples, you may be interested to learn that, per capita, cheese consumption correlates with the number of people dying in any given year by becoming entangled in their bedsheets (r=95%), and the divorce rate in Maine correlates with per capita consumption of margarine (r=99%). Following some of our own analysis, we managed to find a strong correlation between the number of people reading New Europe each week and UK GDP, proving beyond doubt that continued readership is the best guard against recession.