There’s an interesting article in today’s Age on the difficulty of long term forecasting. This is an area into which I won’t venture as it requires a strong belief in the idea that the future is just like the observed past.
I am willing to make that assumption in some contexts, and for short-term forecasts, but not in all contexts, and never for long-term forecasts. The article cites the work of Nassim Nicholas Taleb, author of the interesting book The Black Swan, which discusses events which appear highly improbable given the observed history (such as the discovery of black swans in Australia by people who had only ever seen white swans). There was a similar piece on long-term energy forecasting in the Australian in late May 2007.
This was the second major forecasting piece in the Age in the last few weeks, which is not bad for a mainstream daily newspaper. The previous piece was on 16 August 2007 and was entitled Playing the numbers game. This made the unsurprising point that individuals are harder to forecast than groups. As Sherlock Holmes said (The Sign of the Four, 1890), “You can, for example, never foretell what any one man will do, but you can say with precision what an average number will be up to.”
And while I’m on the subject of forecasting in the popular media, I also noticed today a piece in the Wall St Journal which discussed some research by our very own Kesten Green. As usual, Kesten is out beating the drum for a structured approach to judgemental forecasting. It’s nice that he’s getting a big audience.
Related Posts:
- How good are economic forecasts?
- Forecasting the recession
- Are we getting better at forecasting?
- Academic citations in the popular press
- Prediction markets
Rob J Hyndman