Is Quant Analysis a passing fad

[IndusGIR]

Hi

I want to know how much of this Quant stuff is really working in making real dollars and how much of it is just plain hype.

Who are the folks making money using these techniques and how do they compare to the ones used by the masters like Graham and his followers.

In Aug of 2007 all these models fell apart big time. It is an irony that LTCM which was itself one of the big Quant proponents is not afloat.

Now my question to you folks is what is the average track record of a portfolio manager using Quant methods? Who are these and what percentage of the total managers are they?


thanks
IndusGIR
INDUS Global Investment Research

 

[Sanket Patel]

As for LTCM, you have to remember that there was also the "human" factor involved. If you look at LTCM devoid of any "human" element, their models(s) were near perfect. In fact, if LTCM had continued to hold their positions, they would have turned out ok. But of course, the margin calls along with bureaucratic nonsense killed the fund, not mention they were screwed big time Goldman Sachs and Bear Stearns.

Graham applied methods which fall into the realm of classical finance - he was a pure fundamentals guy. Quant work doesn't quite fit that mold.

Perhaps Dominic can give you a more direct, straightforward answer.

 

[IndusGIR]

Your comment reminds me of some words of wisdom from John Maynard Keynes , "Markets can remain irrational longer than you can remain solvent." I think that answers your point " if LTCM had continued to hold their positions"

 

[Sanket Patel]

I agree with what Keynes said but what I'm saying is that LTCM's collapse was not exclusively a function of LTCM's trading activity. Read "When Genius Failed" by Roger Lowenstein. It seems you're rather wary of anything that is beyond the realm of classical finance.

Does your company happen to have website? I searched Google and I was able to find is your blog. Any particular reason why there isn't a website? I'd like a know bit more about your company's investment philisophy. Perhaps you could elucidate a bit on what distinguishes your company's methodology from simply holding the market index.

 

[IndusGIR]

I do have a lot of respect for classical finance but we test our own beliefs all the time here. In fact we use a combination of different techniques.


My question comes to this forum mainly because I could not find the numbers about any solid results coming from the Quant Portfolio Managers as I have seen from the Conventional Portfolio Managers. On the contrary, I have seen more failures coming from Quant side as pointed earlier.


Regarding your question of holding the index, we know a lot of new people ask that.

What we say is that the more important thing about holding an index (or for that matter any position) is to know when to sell it and what markets you want to hold. That is the key challenge.


If you just held the US Diamonds you would be sitting on even money finally after eight years. That means negative returns in terms of real dollars.

 

[DominiConnor]

I think the thing to understand here is that **** happens.
Nothing actually works.
... all the time.

I counsel you to read http://www.darwinawards.com/ the Dawin Awards (with apologies to any Creationsts around here).

In it you will find people able to misuse toasters, cars, hammers, and pretty much ever other artefact of the human race in such a way as to get themselves removed from the gene pool.

Imagine that QF actually worked, 100%, what would happen ?
It's like the first atomic bomb, which entertainingly used maths many here would recognise from classical quant finance.
It was suspected by some that it would generate so much heat that it would ignite the hydrogen with a fusion reaction that would sterilise the Earth.
...if it worked with 100% efficiency.

It tells you something about the mindset of the time&place that the first atomic bomb was detonated near the middle of America....

You would not need all that good a model to get pretty much all the money in the world.

The ability to forecast, rather than merely fit the observed data points is what any scientist would call the true test of any model or analysis. A weatherman who tells you it rained yesterday, or a doctor who knows the cause of death isn't really that useful.

Forecast almost any financial quantity with the sort of accuracy expected in other disciplines and you would become stupidly rich very quickly.

So basically we're near the start of this journey. Economics as a proper science is only slightly older than some of the people who do it. The idea of sucking in maths and computer techniques is very new.

Many financial techniques have the same characteristics as the early computers. They are big, fragile and hugely expensive, only the domain of big outfits.

One smallish non-financial outfit I know cannot find a way to hedge it's currency exposure at any sensible cost. Hedging $/£ is something we all regard as trivial, but in an globalised economy many firms aren't able to operate properly because they're not big enough to manage FX risk.
That's just one example, finance is far from a mature industry.