L
Liquid validity
Relax, I agree with you on that score, its nothing new, I know that.
At the risk of repeating myself, I'll say it again, they are exacerbating
the issue, not necessarily causing it.
Trade what you see, not what you think....
- Thread starter robster970
- Start date
- Watchers 11
At the risk of repeating myself, I'll say it again, they are exacerbating
the issue, not necessarily causing it.
new_trader
Legendary member
- Messages
- 6,770
- Likes
- 1,655
OK, but that isn't irrefutable fact as you claim. It is just your opinion.
wackypete2
Legendary member
- Messages
- 10,211
- Likes
- 2,058
Sorry, Joe, but the brute force search was on 1000's of Kasparov's previous matches all programmed into DB's database. Sure there's much more to it than that but you don't understand the experiment nor the logic of the whole Deep Blue project.
Yes, I have been out of IT for a while...thanks for reminding me I am getting old
Peter
Last edited:
wackypete2
Legendary member
- Messages
- 10,211
- Likes
- 2,058
No argument there. I just don't think you can compare the current state of trading with the Deep Blue project. Different objectives.
Peter
L
Liquid validity
Fair point, there is no evidence to separate the influence of HFT
and the influence of sub prime:
Yes, The Stock Market Has Been This Volatile Before | Ronald Griess | FINANCIAL SENSE
All I will say is we haven't actually had a full on collapse as in 1929,
yet volatility is similar.
That is the main reason I think HFT is contributing to increased volatility.
As you say though, there is no cast iron evidence to support that,
just as there is none to support what you say.
We are both expressing an opinion, thats all.
random12345
Established member
- Messages
- 793
- Likes
- 280
It was an arbitrary point meant to counter the silly obvious comparisons about what computers can't do being made on the human side of the argument. You're right that I shouldn't have made it as it detracted from the first paragraph.
tar
Legendary member
- Messages
- 10,443
- Likes
- 1,314
BTW there is a big difference between automating trading and making a consistent profitable directional-trading automated system the latter is impossible ! They may auto trade as they like and they may try to exploit the markets but that doesn't mean they will stay profitable on the long run . case in point : Here is the performance of Math geeks :cheesy:
Attachments
L
Liquid validity
LTCM were involved in arb not directional.
http://en.wikipedia.org/wiki/Long-Term_Capital_Management
Its all relative, Rentec are also math geeks and do quite well...
Renaissance Technologies' Medallion Fund: Performance Numbers Illustrated ~ market folly
Its not about which is best, not for me anyway.
My view has always been - do whatever suits you.
tar
Legendary member
- Messages
- 10,443
- Likes
- 1,314
If it was arb and they've failed miserably how if it was directional ?!
L
Liquid validity
No guarantees with anything, anyone can fail.
Whats the difference between LTCM and Rentec.
Well if I knew that I wouldn't be here
Seriously, I spose its down to better assessment and management of risk.
Sounds simple in hindsight.
BTW rentec medallion link is borked, here is a better one:
Best Hedge Funds: Jim Simons Medallion Fund?s Returns and Alpha - Insider Monkey
Shakone
Senior member
- Messages
- 2,460
- Likes
- 665
What is the % of volume that is algo/HFT in FX and on S&P? Does anyone have any figures?
Decades ago, or 1929 for New Trader, it would be 0%. So I think we can safely assume it has increased. Why would it increase if humans are better? Do you think it is just to reduce costs and not have to pay big bonuses? If so, then why are buildings being bought by trading companies right next to exchanges, even though the difference in speed would be milliseconds? Why are all these algos being developed at a huge cost of technology, programmers researchers etc, when just one good trader would suffice?
I think some common sense should be applied, that there is a reason why all of this money has been invested in these projects by banks and hedge funds. Whether that changes the market significantly from 1929, or 2005 is a separate issue.
P.S. tar, I don't understand how you can claim it is impossible to find an automated directional trading strategy. You claim if there was one, then everyone would jump on it and it would disappear. But there are 2 problems with that conclusion. What if there are 100 such strategies, which one do you jump on? What if there are 1million such strategies? And secondly what makes you assume everyone can find them? Who can? The argument is flawed, as you could make the same argument for discretionary, and even arbitrage strategies.
Decades ago, or 1929 for New Trader, it would be 0%. So I think we can safely assume it has increased. Why would it increase if humans are better? Do you think it is just to reduce costs and not have to pay big bonuses? If so, then why are buildings being bought by trading companies right next to exchanges, even though the difference in speed would be milliseconds? Why are all these algos being developed at a huge cost of technology, programmers researchers etc, when just one good trader would suffice?
I think some common sense should be applied, that there is a reason why all of this money has been invested in these projects by banks and hedge funds. Whether that changes the market significantly from 1929, or 2005 is a separate issue.
P.S. tar, I don't understand how you can claim it is impossible to find an automated directional trading strategy. You claim if there was one, then everyone would jump on it and it would disappear. But there are 2 problems with that conclusion. What if there are 100 such strategies, which one do you jump on? What if there are 1million such strategies? And secondly what makes you assume everyone can find them? Who can? The argument is flawed, as you could make the same argument for discretionary, and even arbitrage strategies.
Last edited:
tar
Legendary member
- Messages
- 10,443
- Likes
- 1,314
No guarantees with anything, anyone can fail.
Whats the difference between LTCM and Rentec.
Well if I knew that I wouldn't be here
Seriously, I spose its down to better assessment and management of risk.
Sounds simple in hindsight.
BTW rentec medallion link is borked, here is a better one:
Best Hedge Funds: Jim Simons Medallion Fund?s Returns and Alpha - Insider Monkey
Does Medallion automate directional trading ?
random12345
Established member
- Messages
- 793
- Likes
- 280
Refer to my post about sampled optimisation.
Refer to my numerous posts about JP Morgan's FX unit. E-mail them and explain your points because our little forex world sure can't afford the collapse of such a liquidity provider. You need to save them asap. They need saving.
Refer to the job page of this Hedge fund that trades only automated DIRECTIONAL systems and has done for 20 years. Their profit participation is almost 40% since their return is so crazy good compared to the market average. https://www.renfund.com/vm/jobs.vm - e-mail them and tell them they've got the whole thing wrong and they should employ you instead.
Do you have any actual experience with advanced automated systems (not EAs or equivalent)? Can you really put your hand on heart and claim to know what you're talking about and that your statement is EMPIRICAL fact?
Why make such statements... why...
It's like me saying manual traders won't stay profitable in the long run because I can quote the miserable declines of numerous big names throughout history (including Livermore...). It's not a factual statement however so I wouldn't make it..
F**k sake.
L
Liquid validity
No hard figures, this is all that currently springs to mind.
I'll post something else as well if I remember where I filed the link :whistling
http://www.conatum.com/presscites/HFTMMI.pdf
Not exactly the in depth figs you're looking for I know.
Last edited:
random12345
Established member
- Messages
- 793
- Likes
- 280
Unless you were making such an insane point about your definition of directional trading that you literally meant 100% predictive powers of algorithms. Why even note such an obvious thing?
new_trader
Legendary member
- Messages
- 6,770
- Likes
- 1,655
Let’s say a large pool of investors, or a hedge fund, want to accumulate shares at a very low price, mark them up, and then dump them on the unsuspecting public. What do you think would be the best way to achieve their goal?
1) Investing in a bank of CRAY supercomputers and paying a team of developers to come up with an algorithm that may or may not work?
or
2) Use the tried and true method of having some ‘Expert’ analysts talk down the stock and publish articles in Newspapers etc. focusing only on the negative aspects while selling the stock short. Then doing the reverse after they have accumulated their line. A pump and dump basically.
Pump and Dump Vs CRAY...I know where I'd be putting my money :cheesy:
tar
Legendary member
- Messages
- 10,443
- Likes
- 1,314
Random i wasn't referring to banks trading as a whole they make profits because they have many divisions and they are diversified , maybe they autotrade thousands of systems simultaneously , not to mention they are market makers as well so that is not the point they make money at least most of them although i wouldn't be surprised if all of them blow up at some point ! I was just referring to the argument that a certain group of inputs can be automated to make money consistently (directional trading ) , that argument is flawed and counterproductive ....
Last edited:
L
Liquid validity
robster970
Veteren member
- Messages
- 4,567
- Likes
- 1,390
Because scaling a trading operation using automation is more profitable than scaling a trading operation using people. I know it was a rhetorical question Shakone.
This is not the same as "Can a human trader consistently outperform an algo"?
