Some quants systems no doubt focus on either one vector or one factor. Some are no doubt multi-factor. While the truly ambitious are no doubt seeking a unified theory of the markets, ie. try and incorporate all known anomalies.
Such a system would incorporate small cap
bias, momentum bias, quality bias, value bias, broad diversification,
red flags.
It is the red flags area where short sellers focus. Myriad changes reflected in a firm's accounts may point to warning signs. You can model those and you can model management quality (based on history), but identifying frauds or accounting shenanigans is an incredibly subjective endeavor when the data points you are working with are inconclusive (and they are always inconclusive).
I guess that is why "fundamental based", qualititative oriented investors are now getting caught with their hands in the cookie jar, ie. insider trading. Finding and getting an edge is extremely difficult.
On a side note. I am amazed when I look at so-called small and micro caps where the marketing material of managers in the space talk about inefficiencies due to lack of coverage, that the stocks are generally fairly valued. Now fairly valued is in the eye of the beholder, but there are few glaring mis-valuations from what I can see across the whole spectrum.
Now back to the quants achilles heel. The rigidity of their systems (whether is a static set of criteria or pattern recognition algorithms) or the backward looking nature of the learning systems mean that quants will always miss the nuance of something and/or miss the change in environment because they lack cognitive awareness. And they will never catch the subjective dimensions of not trusting management that comes from a visceral gut feeling.
No comments:
Post a Comment