Thursday, March 22, 2018

Forecasts are often biased - So do it yourself and be held accountable

Forecasts may tell you a great deal about the forecaster; they tell you nothing about the future. - Warren Buffett

There are forecast biases from any opinions sold on the street. Investors should expect biases and you are getting the opinions of an individual or group of individuals which are not fully accountable for the view they sell.

Some of these biases are from herding and trying to be close to the market consensus. No forecaster wants to be an outlier and have a career ender. Some biases are from the strategic behavior of the forecaster who is trying to either preserve or grow a reputation. It could also be caused by optimism or pessimism by the forecaster or irrationality, the lack of using all information available.  These biases can lead to asymmetric loss behavior which is hard to detect but has real effects on the users of the forecast.

The critical issue is whether the forecasters have "skin in the game", the current theme of Nassim Taleb. The best way for a manager or investor to have skin in the game is to develop their own forecasts and be held accountable for what the objective of the forecast. 

There are limited excuses for failure if the decision-maker is also the forecaster. The question is clear - did my forecast make money? An advantage of systematic (quant) fund managers is that they place their forecasting skill directly in the their program. The inputs of any forecast are perfectly known. There is a direct link between forecast quality and dollars made. Even trend-followers, which try to exploit price direction, are held accountable for their "forecasts" through the decisions and profits made. Forecasting bias may not be eliminated but the forecasts is held accountable. 

No comments: