Quants --Alive and Kicking

[Rocko1]

Quote from TSGannGalt:

No one can.

The basic premises of any model or trade is "If the current condition sustains...

... the prices can move up to xxxxx.
... the trend will continue.
... the market will reverse.
... the system will make $xxxx.
... the system will return $xxxx.

etc.
etc.

The difference is how you define "condition" and the conditions change. It can be trend... it can be tendency... it can economical condition... it can patterns... it can be players.... order flow.... it can be anything about the market.

It's the "market" part of "Markets always change"

What you just said takes no more than an understanding of basic algebra to apply. I've met a lot of these so-called quants who understand very little of advanced mathematics.

Real mathematicians are not so simple minded to consider markets stationary or in any linear fashion. There's a lot of research with very promising potential in financial forecasting with concepts such as support vector regression, Kalman filter, and etc.

I've met a lot of other math majors in college who're simply incapable of thinking outside the box, and they end up hurting reputation of financial engineering. It's like any other industry, you have a few winners who stay low profile, while a whole lot of losers get publicity.

 

[TraderZones]

Quote from JR-Welling:

A number of the quant models worked off the assumption that the price of housing never went down.

On that note I think it should become a standard requirement that any person who is majoring in quantitative finance to take a class in common sense 101.

Why does it take hundreds to thousands of people to come up with a limited set of (math) rules to unsucessfully model risk, as with this year's catastrophe?

Rule 1: the future may differ from the past. For example, housing prices don't always go up in the same way that a few years ago, people learned that the stock market does NOT always go up 20% per year

Rule 2: If you think the future may always be the same, see rule number 1

 

[science_trader]

Quote from Thunderdog:

Nothwithstanding some of the hot air that gets passed around in these here parts, I suspect that the underlying physics and chemistry of gases may not be quite as directly applicable to the markets as some would like to believe. Unlike you, I'm not a man of science, so I can't speak with any authority on the matter. However, I suspect that human behavior is a tad more indeterminate and perhaps better analyzed with far more blunt instruments. But if you have made your fortune in the markets using your knowledge of physics, then who am I to question the emperor's clothing?

You don't need to modelize the way people take their decisions. The first level model starts once an individual takes a decision to buy or sell. Then looking at the aggregation of these decisions you can have a fair idea of where the herd is going and if the trend will continue or reverse. The second level model is the feedback effect from the herd to the individual. A bit more complicated but still workable.

Where to get this basic piece of information? orderbooks!

Thermodynamics is the same. You can find empirical law which work very well but still can find exactly the same results starting from individual interactions between molecules. Then a plain N->infinity with some ratio kept constant and here you are. Humans are not idfferent (once their decision is taken and their behavior under certain conditions fairly reproductible).

 

[ML_QUANT]

Predicting the Markets is no more complicated than counting the number of Sands on a Beach!
If you can do one, then you can do the other too...

 

[Euler]

Quote from nitro:



"In the short run, we will learn something. I the intermediate term, we will learn quite a bit. In the long run, we will learn nothing" - ?

Heh! Good one. Not always true, but often enough, especially when it comes to investing manias.

 

[bighog]

All the science, math etc so called whiz kids always seem to forget about the "BUTTERFLY EFFECT".

Nitro is way off base even thinking 20 year old college kids run the show when not popping their zits.

http://answers.yahoo.com/question/index?qid=20080930020838AA9oZiA

There is an age old saying about the mkts for predicting.............. You can tell them a price and you can give them a time but you never tell them both.

PS: thats always been my thinking about trading, the smaller time frames are the easy ones to trade (5 minute charts, ES) because within that time frame the players are reacting on their emotions more than knowledge and i can exploit the situation only because i have a handle on what is happening in the here and now.

 

[FightTheFuture]

Quote from JR-Welling:

A number of the quant models worked off the assumption that the price of housing never went down.

On that note I think it should become a standard requirement that any person who is majoring in quantitative finance to take a class in common sense 101.

I'm afraid they would still fail common sense 101, but the professors would still give them a passing grade just to get rid of them.

Just two final exam true or false questions for quants:

Price only goes up?

A 3% drop in price will not wipe you out when you are leveraged 33:1? (Bear Stearns)

 

[WallstYouth]

Quote from FightTheFuture:

I'm afraid they would still fail common sense 101, but the professors would still give them a passing grade just to get rid of them.

Just two final exam true or false questions for quants:

Price only goes up?

A 3% drop in price will not wipe you out when you are leveraged 33:1? (Bear Stearns)

You my friend are retard as just about ever post in this thread. Quant are the future and are a lot smarter than many could even relish here on ET.

 

[pepper_john]

I know many mortgage related models, but I don't recall any one that assumes that the home prices only go up.

 

[WallstYouth]

Quote from pepper_john:

I know many mortgage related models, but I don't recall any one that assumes that the home prices only go up.

Do you actually believe this crap about quant writing models where housing prices only go up??? At the rate of which prices whet down is what destroyed the system and the fact that many quants implemented the same strategies didnt help the situation.