Consider this a teaser about a big announcement coming next week:
It seems that a certain high end quant tool is going to be made available to the public. I am working to see what we can do for blog readers.
Ultra-luxe, very powerful, and absurdly reasonably priced.
Its gonna be way way cool.
It’s that lame dartboard again, isn’t it?
Fusion IQ for individual investor?
Obviously you haven’t seen the latest CPI ex-dartboards, ex-peanuts, ex-monkeys…
Serious for a moment. Quant tools are almost always “Anti-Coase.” That means they exploit time or information lags. Coase’s fatal flaw is in assuming perfect knowledge perfectly disseminated. With respect to quants dissemination reduces effectiveness. The only conclusion is that the model has leaked and they have a new model that takes this into account.
Barry,
Is this the quant tool that you’ve been developing?
Great so now what has steered the pro’s wrong can steer the amateur’s wrong as well.
Hooray for progress!
Cool. Sumpin we can fade ? LOL
Please, change my world – I’ve used/test-driven just about all of them – I hope it is good!
The pros who went wrong iused a very specific paired trade that had both side go awry.
This is a simpler ranking system (1-100), with each decile yielding better subsequent average performance as a group.
Then a layer of proprietary algorithms get layered over for specific types of screens — groups, sectors, lots of other signals.
More to come . . .
I’m a statistical person, I’m curious. But I’m a Bayesian, the frequentists are ruining the world, they don’t know how to keep it on the road.
well, i’l believe it when i see it. if it consistently outperforms over a long period of time without huge drawdowns, that will satisfy me.
however, i think objective human thought and logic, accounting for market psychology as well as fundamentals, is the only way to perform over time, the market is too complicated a beast to tame with a model.
Consider that even the solution to the game of chess hasn’t been done yet due to the number of potential positions on the board (roughly the same as the number of atoms in the earth!), then i don’t see how you can take account of all the inputs in the market, which are many many orders of magnitude higher than a simple game of chess.
Bring it on, Barry. I will look at it.
Better to either talk to a chess grand master or a physicist before making statements such as this.Chess has been conquered by a computer, many many years ago, now if you had used a card game as an example it would have been a fair comparison.
Um, Michael Butler, try again. What 2and20 said about chess was roughly accurate. You seem to be thinking about Deep Blue or one of the systems that beat Kasparov et al.; but creating a machine that can conquer the best human of our time is nothing like conquering the game itself. The full decision tree of chess will never be laid out by any computer, unless something like quantum computing increases memory/computing power beyond our straightforward physical limits.
But, you are correct that most card games are far more computationally tractable than chess.
Mathematician, thanks.
Micheal Butler, i AM a physicist..although not even close to being a chess grand master :)
Quant Tools – Used in Investments
A scientific phrase – using Quant Tools – will become a common phrase in thwarting a decline in markets – by paralleling two separate types of investments to. Investors will be targeted by fund managers (who are looking for progressive fees.) – Watch out! A few managers have become successful. An example of this type of investment. Two years ago AMD was selling for $40 and INTC (intel) was selling for $18. Buying Intel and selling short AMD (at the same time) would have rewarded you a $10 profit in Intel (18 to 28) and a $30 prolfit in AMD – short (40 to 10.)
Intel and AMD are the only competitors in a certain chip type of chip – while one was going to recover and the other was going to decline – if they are picked and “paired” in an investment – you can make money on both! This was a classic “pairing” – Many situations can be “paired” in investments – for profit in a declining market. – from
http://www.peachin.blogspot.com
I recently dumped AMD for INTC when I saw that INTC has three times as much cash on hand as AMD has market cap and INTC is as vindictive as MSFT. The second fact has much more weighting in my trade decision than the first. Now why AMD’s CEO chose to pick a pricing fight with the biggest karate expert at the bar maybe he can tell us when he gets fired. That being said I am much happier knowing I preserved capital by getting on board the much safer chip train
It will most likely confuse most users, and on top of that, if everyone is using this new ‘quant tool’ what benefit will it have to anyone?
Just look at the machine trades the past 2 weeks, everything in lockstep.
You’d be better of throwing darts.
IBM’s multiple chess computing projects were nothing more than hard coding the moves of the great chess champions into a computer. There is zero intelligence involved. The speed of the systems simply allowed more and more hard coded moves to be analyzed before making the next step. So, the concept of a computer beating a human in chess is absolutely ludicrous. There is no such thing as artificial intelligence. NO SUCH THING. At least not the way lay people view computer intelligence.
And, I respectfully disagree that it was specific pairs trading or specific models that created this mess. Quantitative investing in general created this mess. A failure to recognize this is a failure to recognize the entirety of the problems that exist.
But, in the hands of someone who is a strong fundamentalist, quantitative tools can be a great decision support tool. In other words, not the rationale for the investment but a confirmation of the fundamentals. And, that is all quantitative tools are. In the right hands, Barry’s project will be beneficial.
BDG123,
*Sigh*. More lay people spouting gibberish. You are absolutely incorrect on two major points. First of all, Deep Blue did not simply “hard-code” the moves of great chess masters–rather, it was looking at localized, truncated decision trees. Do a google search on ‘deep blue algorithm’ and learn about the wonders of minimum-window alpha-beta search….
Second of all, Deep Blue definitely did ‘beat’ humans at chess. ‘Winning’ has a clear meaning within the game of chess, and isn’t related to your vehement yet ignorant interpretation.
All that said, I mostly agree that artificial intelligence (of the kind that lay people envision) simply does not exist, for now. Looks like you got lucky, son.
I hope it’s “Spin the Bottle”. I always like to get kissed before I get fu**ed.
You can call it whatever you want to call it. The capability of humans was coded into the machine. A computer did not beat humans at chess. And, there is no such thing as artificial intelligence.