This article was written by Julien Serbit, portfolio manager at Geneva-based Prime Partners.
Alpha generation is not an exact science, and many investment professionals will admit that outperforming the market consistently is more of an art than a science.
Yet investors are still looking for an answer to the big question: can we forecast the market and outperform consistently?
The quest for alpha
Predicting the future has always fascinated humans. Predictably, finding a way to forecast markets is a very exciting prospect for every investor, even if performance cannot be generated by 'fortune telling'.
This pushes people to look to the past to predict the future. It seems wise to look at what happened before and to learn from the mistakes we made to avoid them next time. Most importantly, only data from the past is available.
Technical analysis, automated trading based on signals or patterns recognition are examples of methodologies where investors hope to find their Holy Grail.
Financial market models are all set with a common objective: extracting future trends based on past evolutions that can allow investors to generate alpha by anticipating the next market move.
All and its opposite?
Now remember one of the most used sentences in finance: 'past performances are not an indicator of future results'.
This may seem odd. Everyone is trying to predict the future on the basis of what happened, and at the same time saying that past results are not a guide of future results. But do not trust appearances: it is very common to do so.
An ability to entertain paradoxes is a key feature of human behavior and not necessarily a negative one. It is just an additional illustration that financials markets are reflecting how we behave, alternating rational phases (with a potential to anticipate) and irrational periods (less predictive and more driven by our emotions).
All in all, the best way to outperform markets repeatedly is to avoid any fortune telling effort and the search for the investor's Holy Grail.
However, combining the analysis of rational market phases - using large amounts of data from the past or powerful calculation tools - and more emotional phases requiring human psychology assessment definitely seems to be an interesting combination.