Random Musings about Forecasting and Decision-Making

The strangest thing about the forecasting world is not that it is a dismal science (which it is) but rather that forecasters share some remarkably primitive biases. Whether you look at purely quantitative forecasts or “expert/guru” forecasts, they have one thing in common: they rarely change their opinions or methods in light of new information. In fact, what I have noticed is that the smarter the person is and the more information they seem to possess, the less likely they are to change their mind.Undoubtedly this is why many genuinely intelligent and knowledgeable experts have blown up large funds or personal trading accounts. Ask a person to give you an opinion on where a market is going, and then notice what happens when the market goes dramatically the other way along with news announcements that seem to conflict with their thesis.  Most of the time this person will tell you that they have not changed their mind, and in fact that it is an even better price to buy (or short). Models or systems suffer from the same problem–they typically do not adjust as conditions or regimes change, nor do they observe their own profitability as a cue. Most economists or technical traders build a framework that assumes continuity and self-similarity of the environment in which a forecast was made in the past.

I spent my early investing experiences as a “value investor” and let me tell you that I learned the hard way many times that the market was more often right than wrong. It was uncanny how well future fundamentals were sometimes “forecasted” by price. At the time, I had no knowledge of technical analysis and lacked the intellectual framework to synthesize a superior decision-making method. Of course, I would ride that “under-valued” stock with a price to book ratio less than 1 all the way to being a penny stock before I gave up. I also sold many of my winners far too early because their P/E indicated they were no longer undervalued. Some of these stocks went on to go up 400% or more, while I was content to make 25% profit.  I did the exact opposite with overvalued stocks or stocks with crappy fundamentals. I was heavily short Fannie Mae and Freddie Mac as well as General Motors in early 2007! Of course,  I got my clock cleaned and got  margin-calls long before they plunged almost to zero. This was a case of being right, but too early to fight the sentiment of the crowd.

Credit these events for waking me up to the world of technical analysis and risk-management. But this still does not address how to create an intellectual framework to make superior decisions or forecasts. It finally dawned on me one day that good forecasting (or decision-making) was a dynamic process involving feedback. In fact, the actual information used to make the initial decisions need not be complex as long as you are willing to adjust after the fact. This is especially true the less money you have at risk at the outset. I learned this principle while playing cards— you can see a lot of hands with a small investment as long as you are willing to fold many times when the situation becomes unfavorable, and stick around only when the situation shows promise as more cards are revealed. In other words, your starting hand criteria can be somewhat loose and include hands that do not have a strong edge provided that you adjust properly to new information as the hand progresses. This sounds somewhat elementary, but revolutionary based on how most experts and average traders/investors make decisions. It means that discretionary traders should pay attention strongly to evidence that invalidates their initial hypothesis and be willing to even reverse their position entirely. It also means that traders should risk little up front, and increase positions as subsequent information confirms their hypothesis.In this manner, you win a lot more when you are right then when you are wrong.

In fact, the whole philosophy of trend following has this general mentality–but what is critically absent is the whole rationale for making the trade in the first place! Trend-followers win for many reasons, but primarily because certain fundamental events and crowd psychology develop and cause prices to move to extremes in a feedback loop that is gradual enough to create a discernible trend. The best combination is to take a fundamental hypothesis about risk and reward and use the trend and risk-management techniques to capitalize on it. This way, you have some sense of risk versus reward in advance that may create a situation with superior payoffs to just the standard edge of a conventional trend entry. Of course there are many ways to improve upon this simple framework–but at least it is a good start. Quantitative forecasting models or systems should take the same tact to create superior performance. More interestingly, a forecasting model need not be highly predictive at all by itself— as long as the errors that it makes are systematic and predictable, it can be actually highly useful. This is where many experts miss the boat– a low model r-squared can appear useless, but actually be far more valuable than a high r-squared model once the adjustment is made to model the sources of error.