Forums - Completely understand why HFT must exist

Forums (http://www.elitetrader.com/vb/index.php)
- Automated Trading (http://www.elitetrader.com/vb/forumdisplay.php?forumid=48)
-- Completely understand why HFT must exist (http://www.elitetrader.com/vb/showthread.php?threadid=233300)

Posted by nitro on 12-22-11 06:58 PM:

Completely understand why HFT must exist

For the first time ever, I completely understand the mathematical reason why HFT and profits at this energy/small times must exist.

It remains to be seen if the understanding at a theoretical level leads to a viable strategy, but for me understanding things on this level helps me connect with other knowledge and usually leads to insights. Interestingly, assuming markets are fractal, I also understand why we see similar oscillations on higher time frames.

__________________
"You have to fix your roof when it's sunny outside" - JFK

Posted by vikana on 12-22-11 07:23 PM:

Come on Nitro ... how about a little detail behind your insight? I could be a bit too much Gløgg this time of year

Happy Holidays and thanks for all your past contributions!

__________________
Free your mind

Posted by nazzdack on 12-22-11 07:41 PM:

Re: Completely understand why HFT must exist

Quote from nitro:
----For the first time ever, I completely understand....
----HFT....
----profits....
----theoretical level....
----viable strategy....
----understanding things....
----knowledge....
----insights.
----markets are fractal....
----similar oscillations....
----higher time frames.

Do you believe that HFT "predicts" the market or "manipulates" the market?

Posted by StarDust9182 on 12-22-11 08:04 PM:

Me too .... its so that people can endlessly talk about it and blame their losses on something outside themselves while re-assuring themselves their gains come from their brilliant trading skill. In short it keeps people in the game who shouldn't be in the game.

I have long maintained that the majority of traders have no idea what they are really trading.

Posted by Random.Capital on 12-22-11 08:24 PM:

HFT is the replacement for nickel and dime spreads.

The rise of HFT is the market's way of signaling that it doesn't work "right" at penny spreads.

Posted by bwolinsky on 12-22-11 10:29 PM:

HFT exists out of market structure, not efficiency.

The motivation of brokers is to seek best execution at all times to earn their commission and to trade against other market participants.

The cure to HFT is to implement a trading environment that is only concerned with accurate fills and elimination of getting picked off by market makers and specialists who know ahead of time how large the book coming down the pipe is. Doing this will eliminate bid/ask spreads entirely, and the existence of pre-determined market hours that generally must always post a best bid and offer is the reason for HFT.

This can be eliminated from the market structure using the technologies I've developed, and will find price better than the transactionalism that is prevalent at every trading company today.

__________________
HOW MUCH IS ENOUGH?

Bud Fox

Wall Street

Posted by nitro on 12-23-11 02:37 PM:

Quote from vikana:

Come on Nitro ... how about a little detail behind your insight? I could be a bit too much Gløgg this time of year

Happy Holidays and thanks for all your past contributions!

Happy Holidays

__________________
"You have to fix your roof when it's sunny outside" - JFK

Posted by Occam on 12-24-11 01:45 AM:

Quote from bwolinsky:

The cure to HFT is to implement a trading environment that is only concerned with accurate fills and elimination of getting picked off by market makers and specialists who know ahead of time how large the book coming down the pipe is. Doing this will eliminate bid/ask spreads entirely, and the existence of pre-determined market hours that generally must always post a best bid and offer is the reason for HFT.

This can be eliminated from the market structure using the technologies I've developed, and will find price better than the transactionalism that is prevalent at every trading company today.

Fischer Black also came up with a system that did not involve market makers, a bid, or an ask. The downside is that you don't know what price you'll get and/or how long your order will take to fill (I don't remember whether it was "and" or "or" in the prior sentence, but that's not important). The problem, of course, is this is not what people want in a liquid capital market.

Your own "technologies" also will have major downsides, probably significant enough to make them less appealing than the status quo to the majority of investors/investees. If you think you've got something as great as you imply, then publish your work in the Journal of Finance or somewhere on the Internet for academic critique.

Posted by savagemp5 on 12-26-11 07:45 AM:

THE LAW OF ONE PRICE.

__________________
Random Walk with EMH ?

Posted by nitro on 12-26-11 06:05 PM:

Quote from savagemp5:

THE LAW OF ONE PRICE.

Well yeah, but that is tautological. I mean on an emergent level.

__________________
"You have to fix your roof when it's sunny outside" - JFK

Posted by bwolinsky on 12-26-11 06:07 PM:

Quote from Occam:

Fischer Black also came up with a system that did not involve market makers, a bid, or an ask. The downside is that you don't know what price you'll get and/or how long your order will take to fill (I don't remember whether it was "and" or "or" in the prior sentence, but that's not important). The problem, of course, is this is not what people want in a liquid capital market.

Your own "technologies" also will have major downsides, probably significant enough to make them less appealing than the status quo to the majority of investors/investees. If you think you've got something as great as you imply, then publish your work in the Journal of Finance or somewhere on the Internet for academic critique.

Fischer Black didn't want to give people a hint about the market price before they traded, and I will.

__________________
HOW MUCH IS ENOUGH?

Bud Fox

Wall Street

Posted by rosy2 on 12-26-11 06:32 PM:

why HFT must exist?

Posted by nitro on 12-26-11 09:07 PM:

Because otherwise P=NP.

__________________
"You have to fix your roof when it's sunny outside" - JFK

Posted by bwolinsky on 12-27-11 12:56 AM:

Quote from nitro:

Because otherwise P=NP.

Profit=net profit?

That's why HFT must exist?

Ummm..... I think there needs to be a little bit more explanation, nitro, maybe try more than 5 sentences than your usual outpost everyone at all costs and make extremely frequent short posts posts.

__________________
HOW MUCH IS ENOUGH?

Bud Fox

Wall Street

Posted by nitro on 12-27-11 01:02 AM:

Quote from bwolinsky:

Profit=net profit?

That's why HFT must exist?

Ummm..... I think there needs to be a little bit more explanation, nitro, maybe try more than 5 sentences than your usual outpost everyone at all costs and make extremely frequent short posts posts.

http://en.wikipedia.org/wiki/P_versus_NP_problem

__________________
"You have to fix your roof when it's sunny outside" - JFK

Posted by bwolinsky on 12-27-11 01:04 AM:

Quote from nitro:

http://en.wikipedia.org/wiki/P_versus_NP_problem

Again, how about 5 sentences of your own that explains the point you're trying to make?

Is it that you are afraid of typing more information than you understand so you don't want to look like an uneducated idiot?

Come on, nitro! At least say something worthwhile than that you are fasciated by subset-sum problems in computer sciences and mathematics.

If you need inspiration, why don't you try the question: what does p=np problems have to do with HFT?

You seem to be trying to make the connection but haven't explained why.

5 sentences, please.

__________________
HOW MUCH IS ENOUGH?

Bud Fox

Wall Street

Posted by kcgoogler on 12-27-11 01:11 AM:

Quote from bwolinsky:

Again, how about 5 sentences of your own that explains the point you're trying to make?

Is it that you are afraid of typing more information than you understand so you don't want to look like an uneducated idiot?

Come on, nitro! At least say something worthwhile than that you are fasciated by subset-sum problems in computer sciences and mathematics.

+1. Good one bwolinsky.

Posted by rosy2 on 12-27-11 02:36 AM:

Quote from kcgoogler:

+1. Good one bwolinsky.

wow, good post. I must have put bwolinsky on ignore by mistake...didn't see it. unfortunatley, this forum doesn't allow changes.

Posted by bwolinsky on 12-27-11 02:42 AM:

Quote from rosy2:

wow, good post. I must have put bwolinsky on ignore by mistake...didn't see it. unfortunatley, this forum doesn't allow changes.

You can change your ignore list at any time.

Ignoring me is like ignoring the professor in the class and you'll never get an A that way.

__________________
HOW MUCH IS ENOUGH?

Bud Fox

Wall Street

Posted by TraDaToR on 12-27-11 04:54 PM:

Quote from bwolinsky:

Ignoring me is like ignoring the professor in the class and you'll never get an A that way.

...

Posted by nitro on 12-28-11 09:50 PM:

Quote from bwolinsky:

Again, how about 5 sentences of your own that explains the point you're trying to make?

Is it that you are afraid of typing more information than you understand so you don't want to look like an uneducated idiot?

Come on, nitro! At least say something worthwhile than that you are fasciated by subset-sum problems in computer sciences and mathematics.

If you need inspiration, why don't you try the question: what does p=np problems have to do with HFT?

You seem to be trying to make the connection but haven't explained why.

5 sentences, please.

__________________
"You have to fix your roof when it's sunny outside" - JFK

Posted by stock777 on 12-28-11 10:10 PM:

wtf

it exists for one simple reason

there are always lots of smart people around who like to make a risk free buck without having to dig ditches

NOT A HFT TRADER

__________________
"Those that know ain't saying, and those saying don't know." - E. A. Neumann

A bear since 1958 and proud of it.

Posted by bwolinsky on 12-28-11 10:24 PM:

Quote from stock777:

wtf

it exists for one simple reason

there are always lots of smart people around who like to make a risk free buck without having to dig ditches

NOT A HFT TRADER

Exactly! WTF!

That's definitely not a HFT Trader, I agree.

Nitro, how's that 5 sentence paragraph coming?

Nitro, this post:

Quote from nitro:

Is not a sentence, so try creating an original literary masterpiece so we'll know more about why P=NP means HFT must exist.

In middleschool the teacher always asked us to answer using part of the question in our sentences, so as long as you start with this:

HFT must exist because...

Let that be your first 4 words, because this thread is pretty hilarious and you won't get out of having to explain the premise of your thread without more explicit explanation(s).

Having found holy grails already, ie:trading systems that make a profit on past price histories alone, I know that markets are not efficient, and n!=np.

There, solved for you?

Your premise is that there aren't any such strategies, when I can tell you with certainty that there are, and even if you don't believe I have one, I guarantee HFT traders have them, but then that means p=np so I will say that p!=np for most, but p=np if your search finds such algorithms and myself having found such algorithms I do know p!=np just as the computer scientists state.

I can dispute every weak EMH because I've solved those problems but everyone has a different method to do this, so it's a case by case basis. The efficiency computer scientists have noticed is not out of the EMH, but out of genetic algorithms, and if you can train your algorithm, then you can train it to be p=np, but most will realize p!=np.

__________________
HOW MUCH IS ENOUGH?

Bud Fox

Wall Street

Posted by bwolinsky on 01-05-12 01:30 AM:

Quote from bwolinsky:

Exactly! WTF!

That's definitely not a HFT Trader, I agree.

Nitro, how's that 5 sentence paragraph coming?

Nitro, this post:

Is not a sentence, so try creating an original literary masterpiece so we'll know more about why P=NP means HFT must exist.

In middleschool the teacher always asked us to answer using part of the question in our sentences, so as long as you start with this:

HFT must exist because...

Let that be your first 4 words, because this thread is pretty hilarious and you won't get out of having to explain the premise of your thread without more explicit explanation(s).

Having found holy grails already, ie:trading systems that make a profit on past price histories alone, I know that markets are not efficient, and n!=np.

There, solved for you?

Your premise is that there aren't any such strategies, when I can tell you with certainty that there are, and even if you don't believe I have one, I guarantee HFT traders have them, but then that means p=np so I will say that p!=np for most, but p=np if your search finds such algorithms and myself having found such algorithms I do know p!=np just as the computer scientists state.

I can dispute every weak EMH because I've solved those problems but everyone has a different method to do this, so it's a case by case basis. The efficiency computer scientists have noticed is not out of the EMH, but out of genetic algorithms, and if you can train your algorithm, then you can train it to be p=np, but most will realize p!=np.

As I was trying to optimize futures folio allocations using my system's daily correlations matrix, my Topology Stanford PhD partner said he was not going to try to solve the Napsack problem, and said he would quit if I asked him to.

__________________
HOW MUCH IS ENOUGH?

Bud Fox

Wall Street

Posted by nitro on 01-16-12 08:43 PM:

Caml Trading from Yaron Minsky on Vimeo.

__________________
"You have to fix your roof when it's sunny outside" - JFK

Posted by StarDust9182 on 01-16-12 09:47 PM:

Quote from bwolinsky:

As I was trying to optimize futures folio allocations using my system's daily correlations matrix, my Topology Stanford PhD partner said he was not going to try to solve the Napsack problem, and said he would quit if I asked him to.

Did he look into using a quantum Turing machine?

Prehaps trading ends when quantum computers appear. Their new HFT algorythm with entangled Cayce computing will place their front running order nanoseconds before you do!

Trading will end with a whimper not a bang.

Posted by denner on 01-16-12 11:07 PM:

A battle between blowhards.

It's no coincidence that both of them might also be clinically insane.

__________________
wealth effect: stock market higher, health care costs higher, unemployment higher, food/energy prices higher, taxes higher, poverty higher, bonuses higher, foreclosures higher, homelessness higher, crime rate higher, bankruptcies higher, unsold cars higher... it's economics 101

Posted by bwolinsky on 01-17-12 12:03 AM:

Quote from denner:

A battle between blowhards.

It's no coincidence that both of them might also be clinically insane.

Battle, blowhards? You solve portfolio allocation optimization using the CAPM. You can't know it for certain, but you'll get approximately close.

The OP, nitro, hasn't contributed more than 20 sentences to this thread, and he's the OP! He doesn't understand what he's talking about, but we don't know what his point is because he hasn't made any attempt to articulate what his understanding is and how that relates to HFT's mandatory existence.

The Napsack problem is efficiency to Computer Scientists, can you solve the algorithm and know you've solved with certainty for the most efficient model, and the EMH in a different sense to Economists is whether you can prove or disprove markets are efficient.

There is no original input. Only a thread meant to boost post count with literally less than 5 sentences of an explanation about his rationalization that he claims to understand HFT must exist because of the Napsack Problem,P=NP, when there's no evidence he does. And if he does realize I've already explained the solution, p!=np in computer science, and p=np for some in the markets but most find p!=np, you have your answer.

__________________
HOW MUCH IS ENOUGH?

Bud Fox

Wall Street

Posted by bwolinsky on 01-17-12 12:07 AM:

Yes, this thread truly is pathetic, nitro.

I count 4 sentences, and that's generous if you want to count "Happy Holidays" as one.

You have chosen to let us think you're stupid, rather than speak, and let us know you are stupid.

__________________
HOW MUCH IS ENOUGH?

Bud Fox

Wall Street

Posted by bwolinsky on 01-17-12 12:31 AM:

I think the point of this thread is to get us to join Vimeo, otherwise the video won't load, and if it's not on youtube, I don't care to watch it.

__________________
HOW MUCH IS ENOUGH?

Bud Fox

Wall Street

Posted by nitro on 01-24-12 08:08 PM:

I did a search on F# and found nothing on ET. I find that interesting. I have decided to dedicated myself to learning it. Seems very straightforward.

let the |> fun begin

__________________
"You have to fix your roof when it's sunny outside" - JFK

Posted by Dialectician on 01-24-12 08:22 PM:

Quote from nitro:

I did a search on F# and found nothing on ET. I find that interesting. I have decided to dedicated myself to learning it. Seems very straightforward.

let the |> fun begin

Yes, it's quite surprising. It's not unknown in the industry, though. CS does all (?) of their modeling in it.

It's very well suited for trading, and a joy to program in. It lends itself nicely to concurrent programming.

Posted by bwolinsky on 01-25-12 12:57 AM:

Quote from nitro:

I did a search on F# and found nothing on ET. I find that interesting. I have decided to dedicated myself to learning it. Seems very straightforward.

let the |> fun begin

All right, 5 sentences that don't explain anything. This is typical nitro.

__________________
HOW MUCH IS ENOUGH?

Bud Fox

Wall Street

Posted by rosy2 on 01-25-12 02:48 AM:

Quote from Dialectician:

Yes, it's quite surprising. It's not unknown in the industry, though. CS does all (?) of their modeling in it.

It's very well suited for trading, and a joy to program in. It lends itself nicely to concurrent programming.

I wouldnt say CS does all there modeling in f#. For the functional crowd, I have seen ocaml, scala, erlang, and haskell used at well known firms. But no one just sits and uses one language anymore.

Posted by Dialectician on 01-25-12 11:34 AM:

Quote from rosy2:

I wouldnt say CS does all there modeling in f#. For the functional crowd, I have seen ocaml, scala, erlang, and haskell used at well known firms. But no one just sits and uses one language anymore.

Yes, I would assume so.

However; they hired a team of F# programmers a few years back, and it is their main modeling language now.

It's because of institutions like CS that F# moved beyond the research level at Microsoft.

http://www.microsoft.com/casestudie...dyid=4000006794

As a mathematician, I find F# very pleasant to use. Sadly, there are many programmers who seem determined to miss out on the joys of functional programming. Apparently too few have read SICP...

Posted by Dialectician on 01-26-12 11:42 AM:

I just want to clarify that I was referring to the GMAG London division of CS, as I was being imprecise earlier.

Posted by nitro on 01-27-12 04:02 PM:

http://fdatamining.blogspot.com/201...in-f-using.html

__________________
"You have to fix your roof when it's sunny outside" - JFK

Posted by bwolinsky on 01-27-12 04:36 PM:

Quote from nitro:

http://fdatamining.blogspot.com/201...in-f-using.html

Do you still not have 5 sentences ready to explain your thread?

He seems to be ignoring me.

__________________
HOW MUCH IS ENOUGH?

Bud Fox

Wall Street

Posted by lorndavies on 01-27-12 11:55 PM:

Quote from bwolinsky:

Do you still not have 5 sentences ready to explain your thread?

He seems to be ignoring me.

it's because you're an ass.

Posted by denner on 01-28-12 12:50 AM:

Quote from bwolinsky:

Do you still not have 5 sentences ready to explain your thread?

He seems to be ignoring me.

Nitro is likely autistic.

Posted by rosy2 on 01-28-12 01:00 AM:

Quote from denner:

Nitro is likely autistic.

my guess... random generator. I will start threads and randomly post to see if theres any difference

Posted by bwolinsky on 01-28-12 01:55 AM:

Quote from lorndavies:

it's because you're an ass.

No, it's more because he doesn't want to seem stupid. I'm not an ass. I know what I'm talking about. He hasn't explained anything because it is so complicated that with the amount of verbiage we have from him shows he really doesn't understand.

Using a problem that the op doesn't know how to solve does not demonstrate knowledgability on the topic, no matter how many random links vaguely related to the topic he'd like to post.

If he'd articulate a hypothesis it would be possible to know, but having 17k posts his alias is in for building post count without much detailed explanation. I've been around elitetrader long enough to know that I'm able to profile those types of posters for their sensationalized thread titles, trolling, and lack of substanative dialogue.

I haven't been rude to him. I would like to see him add to the thread his own thoughts, without which (original dialogue by the OP) there is no real point to creating the thread unless he really had something to say about the napsack problem. The Napsack Problem is related to efficiency in various fields, but since we don't know what about P=NP makes that the reason for HFT's existence this thread will probably always take the long road to nowhere.

__________________
HOW MUCH IS ENOUGH?

Bud Fox

Wall Street

Posted by bwolinsky on 01-28-12 02:15 AM:

Quote from denner:

Nitro is likely autistic.

It's odd. Autism or any mental disorder may be an explanation, but if he really wanted to make a point I wish he'd just come out with it, and engage his readers with reasonable attempts at explanation.

Am I being too harsh? I can't be harsh because I haven't criticized him. There's nothing to criticize. He starts a thread and makes every post either a link or less than 5 sentences when this topic is so pervasive that if we really did know what his rationale is this thread could continue with decent discussion.

I've put my explanation out there. P=np to some, P!=np to most, but I don't know how to prove that theorum mathematically and I don't expect him to either. This isn't a journal or even a formal setting for such explanation, so why doesn't he say something without giving a question for an answer?

If he would just talk generally about the topic than providing background info those of us who really do understand the problem are at a loss for why that means HFT must exist.

P=NP does not answer the question why because it is a problem, and the problem he's using as his explanation doesn't have a solution and certainly not a closed form solution. I've posited my perspective, and now I'd like to hear his. The first couple videos may have been informative to the unacquainted reader, but being educated about it you cannot answer a question with a question because a question is not an answer unless the OP would like to provide an explanation of his answer.

It's this explanation I'm after, and I've tried. I don't know what makes him think he's contributed to the readers understanding of this issue, because I haven't seen more than a few sentences in the entire thread. I've written more than him, but he doesn't have to be intimidated or think I'm trying to be critical, I'd just like to know what he's thinking about this than maybe beginning an educational thread for some other purpose also not explained here.

__________________
HOW MUCH IS ENOUGH?

Bud Fox

Wall Street

Posted by nitro on 02-17-12 08:17 PM:

Mathematical Logic Finds Unexpected Application on Wall Street

In an unexpected development for the depressed market for mathematical logicians, Wall Street has begun quietly and aggressively recruiting proof theorists and recursion theorists for their expertise in applying ordinal notations and ordinal collapsing functions to high-frequency algorithmic trading. An ordinal notation system is used to name each ordinal in a certain initial subsequence of the countable ordinals; such systems have recently been applied by elite trading operations to the parameterization of families of trading strategies of breathtaking sophistication. Ordinal notation high-frequency trading algorithms, also called ordinal arbitrage systems, pit their strategies against similar algorithmic opponents on electronic exchanges for a few fleeting seconds, during which thousands of trades are executed, including exploratory trades that test the strategies of opposing human and machine traders.

The monetary advantage of the current strategy is rapidly exhausted after a lifetime of approximately four seconds–an eternity for a machine, but barely enough time for a human to begin to comprehend what happened. The algorithm then switches to another trading strategy of higher ordinal rank, and uses this for a few seconds on one or more electronic exchanges, and so on, while opponent algorithms attempt the same maneuvers, risking billions of dollars in the process.

The elusive and highly coveted positions for proof theorists on Wall Street, where they are known as trans-quantitative analysts, have not been advertised, to the chagrin of executive recruiters who work on commission. Elite hedge funds and bank holding companies have been discreetly approaching mathematical logicians who have programming experience and who are familiar with arcane software such as the ordinal calculator. A few logicians were offered seven figure salaries, according to a source who was not authorized to speak on the matter.

http://christianmarks.wordpress.com...on-wall-street/

__________________
"You have to fix your roof when it's sunny outside" - JFK

Posted by bwolinsky on 02-17-12 09:36 PM:

Quote from nitro:

Mathematical Logic Finds Unexpected Application on Wall Street

In an unexpected development for the depressed market for mathematical logicians, Wall Street has begun quietly and aggressively recruiting proof theorists and recursion theorists for their expertise in applying ordinal notations and ordinal collapsing functions to high-frequency algorithmic trading. An ordinal notation system is used to name each ordinal in a certain initial subsequence of the countable ordinals; such systems have recently been applied by elite trading operations to the parameterization of families of trading strategies of breathtaking sophistication. Ordinal notation high-frequency trading algorithms, also called ordinal arbitrage systems, pit their strategies against similar algorithmic opponents on electronic exchanges for a few fleeting seconds, during which thousands of trades are executed, including exploratory trades that test the strategies of opposing human and machine traders.

The monetary advantage of the current strategy is rapidly exhausted after a lifetime of approximately four seconds–an eternity for a machine, but barely enough time for a human to begin to comprehend what happened. The algorithm then switches to another trading strategy of higher ordinal rank, and uses this for a few seconds on one or more electronic exchanges, and so on, while opponent algorithms attempt the same maneuvers, risking billions of dollars in the process.

The elusive and highly coveted positions for proof theorists on Wall Street, where they are known as trans-quantitative analysts, have not been advertised, to the chagrin of executive recruiters who work on commission. Elite hedge funds and bank holding companies have been discreetly approaching mathematical logicians who have programming experience and who are familiar with arcane software such as the ordinal calculator. A few logicians were offered seven figure salaries, according to a source who was not authorized to speak on the matter.

http://christianmarks.wordpress.com...on-wall-street/

Do you have anything of your own to say?

__________________
HOW MUCH IS ENOUGH?

Bud Fox

Wall Street

Posted by tenthousandmen on 02-17-12 10:15 PM:

Re: Completely understand why HFT must exist

Quote from nitro:

For the first time ever, I completely understand the mathematical reason why HFT and profits at this energy/small times must exist.

It remains to be seen if the understanding at a theoretical level leads to a viable strategy, but for me understanding things on this level helps me connect with other knowledge and usually leads to insights. Interestingly, assuming markets are fractal, I also understand why we see similar oscillations on higher time frames.

The answer is so you pikers have something to blame

Posted by lorndavies on 02-20-12 09:31 PM:

Quote from bwolinsky:

I'm not an ass. I know what I'm talking about

you think you know what you're talking about, which is what makes you an ass. having a violent hard-on for someone who's putting up great links, also qualifies you for the title.

for instance, that single jane's street link has more value than every word you've probably ever uttered. that he's not anally following some subjective rules you might have about how to conduct a thread is your deal and really shouldn't be anyone else's. so, why don't you let him do his thing, and let the adults go through his links like they were meant to.

Posted by bwolinsky on 02-20-12 09:43 PM:

Quote from lorndavies:

you think you know what you're talking about, which is what makes you an ass. having a violent hard-on for someone who's putting up great links, also qualifies you for the title.

for instance, that single jane's street link has more value than every word you've probably ever uttered. that he's not anally following some subjective rules you might have about how to conduct a thread is your deal and really shouldn't be anyone else's. so, why don't you let him do his thing, and let the adults go through his links like they were meant to.

If you can stand unoriginal content, and see that he's doing it just to get a high post count to get more views, it's really annoying.

I'd like him to explain his thread, than just answering the thread's premise with a question. No one else has offered as detailed an explanation on this subject, not even the OP.

Posting links with no explanation is the same as ignorance. That he might be building up to something for his research is pretty far fetched, but what I really don't like is he can't even say 5 sentences for this thread. If he did that, I'd be fine. His alias has numerous posts with little or no substance, and this is not the sign of a knowledeable poster, but one who is scared for people to judge his intellectual capacity when I'd like him to write something original.

__________________
HOW MUCH IS ENOUGH?

Bud Fox

Wall Street

Posted by maxpi on 02-23-12 01:49 PM:

Quote from denner:

A battle between blowhards.

It's no coincidence that both of them might also be clinically insane.

minor details! What's a little insanity among ET'ers... I'm wondering who makes more money, and I'm betting on the latter: those two or the typical successful trader that has no clue what either is talking about...

Posted by eusdaiki on 02-23-12 08:11 PM:

@nitro

That Janestreet video was pretty cool... thnx.

@bwolinsky

The ignore feature works wonderfully to remove annoying people from your screen

On the other hand, I do tend to agree with your request... it'll be nice to see an explanation on Nitro's epiphany...

__________________
We will not have any more crashes in our time."
- John Maynard Keynes in 1927

Posted by bwolinsky on 02-23-12 11:06 PM:

Quote from eusdaiki:

@nitro

That Janestreet video was pretty cool... thnx.

@bwolinsky

The ignore feature works wonderfully to remove annoying people from your screen

On the other hand, I do tend to agree with your request... it'll be nice to see an explanation on Nitro's epiphany...

I only ignore people who attack me personally, or engage in the same debates about my past incessantly. Nitro didn't do that, but, still, what's wrong with asking like the second poster moderator ask him to explain such a profoundly difficult question?

__________________
HOW MUCH IS ENOUGH?

Bud Fox

Wall Street

Posted by highballs on 02-23-12 11:07 PM:

Quote from bwolinsky:

I only ignore people who attack me personally, or engage in the same debates about my past incessantly. Nitro didn't do that, but, still, what's wrong with asking like the second poster moderator ask him to explain such a profoundly difficult question?

nothing wrong with it at all..love the way you call the posers out! keep up the great work!
i subscribe to you!
iread everything!

Posted by nitro on 02-25-12 02:13 PM:

Fractional Brownian motion is a semimartingale only if H = 1/2.

__________________
"You have to fix your roof when it's sunny outside" - JFK

Posted by nitro on 06-04-12 07:41 PM:

I am developing/engineering a distributed high frequency system that cannot tolerate latency and has huge demands [network/computationally] put on it. The level of detail is intense.

Maybe I will write a book about it...

__________________
"You have to fix your roof when it's sunny outside" - JFK

Posted by eusdaiki on 06-04-12 07:50 PM:

Quote from nitro:

Fractional Brownian motion is a semimartingale only if H = 1/2.

Was this your epiphany?

__________________
We will not have any more crashes in our time."
- John Maynard Keynes in 1927

Posted by johnkurtz on 06-04-12 07:53 PM:

Quote from nitro:

I am developing/engineering a distributed high frequency system that cannot tolerate latency and has huge demands [network/computationally] put on it. The level of detail is intense.

Maybe I will write a book about it...

sounds beutiful man.

if it makes CASH $ even prettier.

Posted by elliots11 on 06-07-12 06:07 AM:

HFT is imo legalized insider trading. The firms know order flow before the little guys.

__________________
Elliotstrade1

Posted by goodgoing on 06-11-12 08:09 AM:

Quote from elliots11:

HFT is imo legalized insider trading. The firms know order flow before the little guys.

What a pity for you! Allthough this is wrong about HFT, locals also knew the order flow before the retail in the past. This is why they paid for a seat at the exchange. Just learn how to trade and stop complaining about HFT. It is here to stay and you must learn how to live with it or do something else. Here is a good blog post about the flocking behavior of HFT robots.

Posted by Lornz on 06-11-12 09:14 AM:

Quote from goodgoing:

Here is a good blog post about the flocking behavior of HFT robots.

That's just generic rambling about a well-known phenomena. I don't see what's good about it?

Posted by goodgoing on 06-11-12 12:58 PM:

Quote from Lornz:

That's just generic rambling about a well-known phenomena. I don't see what's good about it?

Flocking is a well-known phenomenon but to be honest I did not know about the analogy to HFT robot flocking. Since it is well-known to you, can you provide some links to this specifically because I am quite interesting in it?

Posted by Lornz on 06-11-12 01:32 PM:

Quote from goodgoing:

Flocking is a well-known phenomenon but to be honest I did not know about the analogy to HFT robot flocking. Since it is well-known to you, can you provide some links to this specifically because I am quite interesting in it?

Are you sure you'd want reading recommendations from an idiot who struggles with the singular form of "phenomenon"?

To be honest, I don't know of any sources that covers it specifically. I've just had an interest in chaos and complexity in general for some time and I've done my own analysis. It's definitely an interesting topic...

Posted by goodgoing on 06-11-12 03:24 PM:

Quote from Lornz:

To be honest, I don't know of any sources that covers it specifically. I've just had an interest in chaos and complexity in general for some time and I've done my own analysis. It's definitely an interesting topic...

You have no references to provide, yet you called the post "ramblings". Why is that? I'm trying to understand. Is the post bad or you are a negative person?

Posted by eusdaiki on 06-11-12 03:31 PM:

Quote from goodgoing:

Flocking is a well-known phenomenon but to be honest I did not know about the analogy to HFT robot flocking. Since it is well-known to you, can you provide some links to this specifically because I am quite interesting in it?

Synchronization (flocking) happens all over the place in nature (even for innanimate objects...)

See this Ted talk for detailed examples:
http://www.youtube.com/watch?v=aSNrKS-sCE0

It is only logical that HFT would follow such universal behaviours...

The Flash crash was a good example of this phenomenon.
this and other illustrations of the phenomenon can be easily observed in the research conducted by nanex

http://www.nanex.net/FlashCrash/OngoingResearch.html

__________________
We will not have any more crashes in our time."
- John Maynard Keynes in 1927

Posted by Lornz on 06-11-12 03:52 PM:

Quote from goodgoing:

You have no references to provide, yet you called the post "ramblings". Why is that? I'm trying to understand. Is the post bad or you are a negative person?

I didn't mean to come off as harsh as I probably did. My criticism was simply that he doesn't provide any evidence for his theory. Where are his references?

He took a topic that's been "everywhere" in popular science the past few years and applied it (conceptually) to HFT. He then states that trading higher time frames might avoid the impact of HFT. These are hardly novel ideas, and I assume every decent trader has thought about both a long time ago. I think I've even seen a few posts about it on this very message board...

The blog post is basically just a few loose thoughts without anything to back them up; not even a theory of how one would exploit the potential occurrence of such a pattern.

Thus, I don't see how the post holds much value to anyone but extremely naive chartists.

Posted by elliots11 on 06-12-12 02:22 AM:

Flocking or chaos theory is not relevant at all to what I said. HFT is akin to legalized front running. The bots see the order flow before our eyes can blink. They are able to trade off the order flow in a rapid manner. The flash crash was not flocking it was a response of algos all triggering sell orders at the same time.

__________________
Elliotstrade1

Posted by eusdaiki on 06-12-12 03:00 AM:

Quote from elliots11:

Flocking or chaos theory is not relevant at all to what I said. HFT is akin to legalized front running. The bots see the order flow before our eyes can blink. They are able to trade off the order flow in a rapid manner. The flash crash was not flocking it was a response of algos all triggering sell orders at the same time.

All the algos were triggering sell orders and cancelling their bids as a response to a large sell order... no different from fish moving out of the way when a dolphin is comming through... once one of the algos starts cancelling the others do the same... just like an stampede...

__________________
We will not have any more crashes in our time."
- John Maynard Keynes in 1927

Posted by elliots11 on 06-12-12 03:16 AM:

Whatever. Still no one has yet to disprove my thesis that HFT is about trading on order flow information not available to those with slower computers or whose gateways are troubled with latency.

__________________
Elliotstrade1

Posted by eusdaiki on 06-12-12 03:33 AM:

Quote from elliots11:

Whatever. Still no one has yet to disprove my thesis that HFT is about trading on order flow information not available to those with slower computers or whose gateways are troubled with latency.

Some hfts do behave as you describe.
Some of them don't.

Because of scale economies they can't all do the same thing while theyre 70% of the volume. They have to do different things.

Some hfts just care about getting a good place in the queue when the price changes to get a better chance at scoring credits.
Some work of complex quant models to try to optimize for an aspect of the markets behavior others aren't looking at.
Behind every hft there's a guy trying to figure out how to take more $ from the market and everyone of them cannot be the fastest ( only one of them actually is) and only a few of them can actually profit from getting info and acting on it before others do. The others have to come up with different ways to make money.

__________________
We will not have any more crashes in our time."
- John Maynard Keynes in 1927

Posted by elliots11 on 06-12-12 03:49 AM:

Thanks for sharing relevant comment on HFT.

__________________
Elliotstrade1

Posted by WinstonTJ on 06-12-12 04:58 AM:

The coined term "HFT" is a joke.

There are a few key factors within automated trading:

Holding Time
Markets traded (or products traded)
Daily Volume
Cancel to Fill Ratios

Latency is also a factor but latency depends more on the above than on it's own.

Posted by elliots11 on 06-12-12 05:02 AM:

payment for providing equity is a part of it as well

__________________
Elliotstrade1

Posted by nitro on 09-29-12 03:38 PM:

Quote from Dialectician:

Yes, I would assume so.

However; they hired a team of F# programmers a few years back, and it is their main modeling language now.

It's because of institutions like CS that F# moved beyond the research level at Microsoft.

http://www.microsoft.com/casestudie...dyid=4000006794

As a mathematician, I find F# very pleasant to use. Sadly, there are many programmers who seem determined to miss out on the joys of functional programming. Apparently too few have read SICP...

SICP

http://sicpebook.files.wordpress.com/2011/11/sicp.pdf

__________________
"You have to fix your roof when it's sunny outside" - JFK

Posted by ddude on 09-29-12 07:10 PM:

Ah yes, SICP, a classic...

Lisp, a dynamic language... FPGAs, dynamic hardware...
Turtles all the way down 8^)

Cordially,
-DD-

"Simplicity is the ultimate sophistication." -Leonardo da Vinci-

Posted by nitro on 10-16-12 06:35 PM:

I found this a bit strange when I read it .But in time it has sunk in and now I understand the idea behind it.

The question is now, are there countably infinite agents, or uncontable number of them. Almost certainly countable...

Quote from nitro:

Mathematical Logic Finds Unexpected Application on Wall Street

In an unexpected development for the depressed market for mathematical logicians, Wall Street has begun quietly and aggressively recruiting proof theorists and recursion theorists for their expertise in applying ordinal notations and ordinal collapsing functions to high-frequency algorithmic trading. An ordinal notation system is used to name each ordinal in a certain initial subsequence of the countable ordinals; such systems have recently been applied by elite trading operations to the parameterization of families of trading strategies of breathtaking sophistication. Ordinal notation high-frequency trading algorithms, also called ordinal arbitrage systems, pit their strategies against similar algorithmic opponents on electronic exchanges for a few fleeting seconds, during which thousands of trades are executed, including exploratory trades that test the strategies of opposing human and machine traders.

The monetary advantage of the current strategy is rapidly exhausted after a lifetime of approximately four seconds–an eternity for a machine, but barely enough time for a human to begin to comprehend what happened. The algorithm then switches to another trading strategy of higher ordinal rank, and uses this for a few seconds on one or more electronic exchanges, and so on, while opponent algorithms attempt the same maneuvers, risking billions of dollars in the process.

The elusive and highly coveted positions for proof theorists on Wall Street, where they are known as trans-quantitative analysts, have not been advertised, to the chagrin of executive recruiters who work on commission. Elite hedge funds and bank holding companies have been discreetly approaching mathematical logicians who have programming experience and who are familiar with arcane software such as the ordinal calculator. A few logicians were offered seven figure salaries, according to a source who was not authorized to speak on the matter.

http://christianmarks.wordpress.com...on-wall-street/

__________________
"You have to fix your roof when it's sunny outside" - JFK

Posted by Random.Capital on 10-16-12 07:01 PM:

While an admirable attempt, what they're trying to do here (which is a bit mis-described in the article, but whatever, close enough) is only achievable in isolation, not in live systems.

Posted by clearinghouse on 10-16-12 07:06 PM:

Quote from nitro:

I found this a bit strange when I read it .But in time it has sunk in and now I understand the idea behind it.

The question is now, are there countably infinite agents, or uncontable number of them. Almost certainly countable...

Kind of cool, but probably a losing bet in terms of salary paid vs. results. Chalk it up to the speculative R&D budget?

Everytime I saw one of these ambitious projects tried in a firm, it always ended up in a layoff a few months later.

But, I'm wililng to believe that there's probably some fly by night shop out there with 3 PhDs (probably all former Soviet defectors who got tired of building nuclear weapons) using this, on an FPGA, with a homebrew microwave network link. They all dress in Walmart clothes and trade out of an RV in New Jersey, and the most we'll ever know about this A-Team of trading is the random obscure blog post talking about their unpublished landmark research.

Oh, and even though they're all super billionaires, their supermodel wives don't tip cab drivers well.

Posted by DeeDeeTwo on 10-16-12 07:17 PM:

Quote from ddude:

"Simplicity is the ultimate sophistication." -Leonardo da Vinci-

Thanks for this quote.

Always known that very smart people make things very complicated...
But geniuses simplify, simplify, simplify.

Very smart people cross the street via the North Pole...
Geniuses find a shortcut... always a shortcut.

Posted by jem on 10-16-12 07:38 PM:

but why does hft have to exist... those profits could have been spread out over a wider collection of people.

As 30 second to 2 minute time frames once allowed me to make a nice living.

It would seem HFT can destablize the market as they are doing battle with each other and blowing each other up. Do they not claim to be providing liquidity?

__________________
"From where Winston stood it was just possible to read, picked out on its white face in elegant lettering, the three slogans of the Party:
WAR IS PEACE
FREEDOM IS SLAVERY
IGNORANCE IS STRENGTH."

Posted by CT10Gov on 10-16-12 07:44 PM:

Why must you and your few friends exist rather than HFT?

HFT merely found a way to more efficiently harvest a niche source of alpha.

One would think on elite 'trader', one doesn't need to be subject to this sort of nonsense about 'distributive justice'.

Quote from jem:

but why does hft have to exist... those profits could have been spread out over a wider collection of people.

As short time frames once allowed me and a few friends to make a nice living.

Posted by jem on 10-16-12 10:08 PM:

its a question of whether we should all be on an even playing field or the field should be tilted to a few with co located servers or faster quotes (which was in the news recently)

I think the playing field should be fair. I do not think crony capitalism is a good model.

Quote from CT10Gov:

Why must you and your few friends exist rather than HFT?

HFT merely found a way to more efficiently harvest a niche source of alpha.

One would think on elite 'trader', one doesn't need to be subject to this sort of nonsense about 'distributive justice'.

Posted by JuniorCTA on 10-16-12 10:11 PM:

Quote from jem:

its a question of whether we should all be on an even playing field or the field should be tilted to a few with co located servers or faster quotes (which was in the news recently)

I think the playing field should be fair. I do not think crony capitalism is a good model.

we should all have ivy league degrees from harvard

Posted by CT10Gov on 10-16-12 10:13 PM:

I agree that we should all detest "crony capitalism". But, the 'crony' part refers to the government. For a devil's advocate's argument: why should a private exchange be able to make rules as it sees fit? Why is it wrong that someone pays the extra mile for the technology and closer server? How fair is fair?

Now, I'm all for everyone should play by the same rule. That being said, only a portion of HFT has to do with the sort of 'absursive' market marketing that everyone likes to trash. A great deal of the others are simply high speed stats arb strategies of sorts.

So it sounds like your problem is with the exchange, not with HFT?

Quote from jem:

its a question of whether we should all be on an even playing field or the field should be tilted to a few with co located servers or faster quotes (which was in the news recently)

I think the playing field should be fair. I do not think crony capitalism is a good model.

Posted by propseeker on 10-16-12 10:38 PM:

Quote from jem:

but why does hft have to exist... those profits could have been spread out over a wider collection of people.

As 30 second to 2 minute time frames once allowed me to make a nice living.

wait, aren't you the guy who spends a good chunk of time in the politics/religion forum bashing the left and pushing conservative/libertarian views?

now you're talking about getting out-competed and think it's a shame that profits aren't being distributed over a wider "collection" of people? ha!

conservative when you're ahead, liberal when you're behind i guess.

Posted by propseeker on 10-16-12 10:47 PM:

Quote from jem:

its a question of whether we should all be on an even playing field or the field should be tilted to a few with co located servers or faster quotes (which was in the news recently)

I think the playing field should be fair. I do not think crony capitalism is a good model.

there's nothing stopping you as a trader in investing in the infrastructure required to compete in the hft world. just like there would be nothing stopping you as a backyard foundryman from investing in the infrastructure to become a major steel producer. there's nothing unfair about getting out-competed, when you don't have the money to compete.

crony capitalism in hft? ummm, where exactly?

Posted by jem on 10-17-12 12:14 AM:

Quote from propseeker:

there's nothing stopping you as a trader in investing in the infrastructure required to compete in the hft world. just like there would be nothing stopping you as a backyard foundryman from investing in the infrastructure to become a major steel producer. there's nothing unfair about getting out-competed, when you don't have the money to compete.

crony capitalism in hft? ummm, where exactly?

I can accept 2 things related to full information...

1. An exchange without the cover of fairness from the SEC and govt.
If it is the wild west let the public know.

or

2. A exchange in which the govt ensures fairness.

What we have here is the false idea of fairness and govt protection. The Cronies have taken SEC, Govt regulators, and Cabinet positions. These posts or filled by banker emeriti or people who know they will be hired by the bankers... Much as generals know they will be defense contractors someday. . Cronies have been getting the benefits and the bailouts. Cronies have had the money to buy such influence because member banks own the fed and the member banks have produced enormous wealth for some of it top bankers.

If the banks and exchanges want to go all private.. then release then eliminate the FED and return all the tax payer dollars from the bailout and all these VA and FHA mortgages.

3. I would even accept full disclosure ahead of time.
What we have seen is HFT people getting quotes early for a few years and then it being disclosed.

What we just saw was apparently some people got better feeds from the NYSE and then later it being disclosed.

4. What I do not like is co located servers.... when I was younger I joined exchanges... now I choose not play. I would prefer an even playing field so I could play again and more of the public might too.

Posted by JuniorCTA on 10-17-12 12:55 AM:

Quote from propseeker:

conservative when you're ahead, liberal when you're behind i guess.

best line i ever read on ET

Posted by Random.Capital on 10-17-12 01:14 AM:

Quote from propseeker:

there's nothing stopping you as a trader in investing in the infrastructure required to compete in the hft world.

In fact, it's never been cheaper or easier to do just that.

Whether it's smart to get involved in that game, at this stage of its evolution, is an entirely different question...

Posted by propseeker on 10-17-12 01:19 AM:

@jem... no one is going to counterpoint you when you think it's ok to broadly and off-topically expand the scope of your argument. in the course of a couple of posts, you've gone from collectivist laments to crony capitalism to fha loans. while i don't doubt your broad and general wisdom, i think it's better suited for showing lefties their place vs trying to educate auto-traders about the intricacies of their business.

Posted by vicirek on 10-17-12 01:35 AM:

Quote from jem:

4. What I do not like is co located servers.... when I was younger I joined exchanges... now I choose not play. I would prefer an even playing field so I could play again and more of the public might too.

All your points 1-3 are valid.

However, how much you or me really loose due to micro - millisecond market player.

Realistically the fastest we can trade is in second time frame without going into more costly infrastructure/overhead. And within that second time frame market price shift should not matter that much because your strategy would have to be based on much longer time frame.

I would say it is not a factor and HFT guy may actually operate at a loss which is your gain - probably about 50% of time.

Markets were never fair and now are a little bit more transparent with new technology.

As to the principle and technology:

Flashing bids/offers on some internal loop and then disseminating them when they are history is kind of playing without risk. They hit only profitable price differences that are visible to them which amounts to being market maker who by definition front runs order flow by matching customer orders.

These guys exploit microstructure of the markets and various speeds at which internal networks operate compared to external loops and play inter-market arbitrage without much risk.

One alternative is to go back to phones and specialists so fund manager can call specialist and fix prices of their holdings.

Another is to simplify rules so they are enforceable and institute central queuing system or tokenization and any quote in the system is controlled by that queue not by local exchange after it is launched. It can be cancelled or modified but at lower priority than executing order. Payments and crossing orders from different exchanges would follow certain predetermined protocol. That way everybody has skin in the game and if somebody wants to slow the system with quotes will automatically slow themselves as well. I guess such system could be gamed as well.

So what is your idea of even playing field and who would provide continuous liquidity and quotation to the markets knowing that they are taking excessive execution risk?

__________________
mk

Posted by propseeker on 10-17-12 02:16 AM:

Quote from vicirek:

As to the principle and technology:

Flashing bids/offers on some internal loop and then disseminating them when they are history is kind of playing without risk. They hit only profitable price differences that are visible to them which amounts to being market maker who by definition front runs order flow by matching customer orders.

this is fantasy, only if one makes an extreme effort to understand it. otherwise, it's gibberish.

Posted by vicirek on 10-17-12 03:04 AM:

Quote from propseeker:

this is fantasy, only if one makes an extreme effort to understand it. otherwise, it's gibberish.

Do something useful and fix it.

__________________
mk

Posted by Random.Capital on 10-17-12 03:36 AM:

Quote from vicirek:

Do something useful and fix it.

It's gibberish because you have the same basic tools as "them", and can easily force them to execute on those "flashed in a loop" orders that seem to be bothering you so much.

Posted by vicirek on 10-17-12 04:56 AM:

Quote from Random.Capital:

It's gibberish because you have the same basic tools as "them", and can easily force them to execute on those "flashed in a loop" orders that seem to be bothering you so much.

Correction: they do not bother me a bit.

I do not think that you can execute on quote that has been cancelled even if I had resting order somewhere - usually it would be in the different segment of the market ( ECN broker exchange ) and regular routing re-routing would not be able to catch it before cancel.

Whatever I see on the screen or my computer senses as a signal is history and as long as I know that it is OK with me.

I care more about tight spread and immediate fill rather than some minor slippage which can be defined as a difference between perceived and actual quote at the time of execution (sometimes is not minor but I can get even with "them").

__________________
mk

Posted by Random.Capital on 10-17-12 05:04 AM:

Quote from vicirek:

I do not think that you can execute on quote that has been cancelled...

You're really not getting it, mate.

Anyway, as it doesn't bother you, no point in going any further.

Cheers.

Posted by vicirek on 10-17-12 05:08 AM:

Quote from Random.Capital:

You're really not getting it, mate.

Anyway, as it doesn't bother you, no point in going any further.

Cheers.

Go ahead. I need some info.

__________________
mk

Posted by nitro on 10-18-12 02:00 PM:

Quote from CT10Gov:

I agree that we should all detest "crony capitalism". But, the 'crony' part refers to the government. For a devil's advocate's argument: why should a private exchange be able to make rules as it sees fit? Why is it wrong that someone pays the extra mile for the technology and closer server? How fair is fair?

Now, I'm all for everyone should play by the same rule. That being said, only a portion of HFT has to do with the sort of 'absursive' market marketing that everyone likes to trash. A great deal of the others are simply high speed stats arb strategies of sorts.

So it sounds like your problem is with the exchange, not with HFT?

I think the idea is similar to that of a track and field event where the race has to go around at least one turn. In that case, "fair" just means that lane 1 starts farther back than lane 2 and so on. Proximity to exchange should similarly be made as fair as possible, IMO.

With the advent of virtual servers this is getting more and more democratic IMO because affording colocation is then within the retail trader means.

__________________
"You have to fix your roof when it's sunny outside" - JFK

Posted by CT10Gov on 10-18-12 02:19 PM:

I get the 'fairness' argument here, but upon reflection, I'm not sure I fully agree that it's necessary for proximity to exchange to be fair.

First, it's never been fair. Recall the days when the ol' time traders would employ their own agents and runners at the ports for information coming off of ships and send them to the agents physically sitting at the exchange? That's colocation in a sense. Proximity to exchange and the asymmetric information has always been a part of the market place. It's not new,

Which brings me to my second point: the market place is suppose to be a mechanism of information discovery. It's not predicated on symmetrically informed participants. In fact, it's great feature is that it doesn't need it. So why force some arbitrary standard of fairness on its participants? Arguably, the disparity between the capital level of different participants is another source of unfairness: should we do something about that too?

Finally, I'm not convinced that all this hurting the market. Sure, old ET-style scalpers are done with - but so what? Why should they be entitled to earn a living. Is Buffet hurt by HFT? Are buy-side institutions more hurt by HFT or helped by the execution algorithms that are encompassed? Does it hurt my 401k investments?

Quote from nitro:

I think the idea is similar to that of a track and field event where the race has to go around at least one turn. In that case, "fair" just means that lane 1 starts farther back than lane 2 and so on. Proximity to exchange should similarly be made as fair as possible, IMO.

Posted by LincolnArmy on 10-18-12 02:34 PM:

The problem with letting a private exchange set its own rules is that we need a capitalist system for this to make sense.

Take the CME Group. This is now a TBTF institution. There are huge barriers to entry in this industry and no effective competition for it's contracts. It has been allowed to buy up any exchange that has offered any competition.

It's hard to back any of the decisions/policies of the CME when it has proved unable to police the clearing firms and safeguard client money (FUCK YOU JON CORZINE).

The arguments backing HFT on the premise that the exchanges operate in a capitalist environment with a competent, rational management are without foundation.

Posted by CT10Gov on 10-18-12 02:52 PM:

Eh, with millions of transactions every business day, there's one a*shole of Corzine's magnitude every ten years or so... it's surely not perfect, but it's not terrible either.

In any case, what are you proposing exactly? Nationalized exchanges?

Quote from LincolnArmy:

It's hard to back any of the decisions/policies of the CME when it has proved unable to police the clearing firms and safeguard client money (FUCK YOU JON CORZINE).

Posted by LincolnArmy on 10-18-12 03:27 PM:

True enough there will always be criminals like Jon Corzine -there was no value to me mentioning his name -I just find it cathartic to say FUCK YOU JON CORZINE at any given opportunity.

I'm not proposing nationalized exchanges. As a capitalist I find that goes against the grain but we have left capitalism so far behind us that it is an option among others that I would consider.
I remember back in the 80s and early 90s in the country where I grew up there used to be a monopolies and mergers commission that mostly stopped any merger that would result in over 20% market share for the newly formed entity.

Economies of scale and synergies after takeovers, abundant in Econ textbooks rarely take place in the real world. I think having the CBOT and CME competing was a good thing and I think it's a shame the members sold out and let the place get taken over by lawyers and accountants.

So many arguments these days are made on the basis of capitalism versus socialism (esp with the election coming up) and this is a false premise given the falacy that what we have today bears any resemblence to capitalism. Look at the LIFFE exchange where RSJ have made claims that if you trade certain products there, then there is an 80% chance that they will be the counterparty -that's a monopoly in my book.

I know it's getting a bit long winded here, but my simple point was that we have wondered so far away from competition and capitalist concepts such as price discovery that to me, to introduce them, as the foundation of an argument, is false.

Posted by CT10Gov on 10-18-12 03:32 PM:

Why? Stock markets and futures markets are still pretty damn good places for information discovery. Look at how the 10-yr future reacts to Fed announcements, or Corn futures react to usda releases.

If markets had no price discovery ability left, then it should have no trading at all, or extremely smooth pricing since everyone already knows everything. Neither is true.

There's an old saying right? It's a good deal if neither sides are happy. Everyone's miserable in the market place - maybe that's a good thing?

Quote from LincolnArmy:

I know it's getting a bit long winded here, but my simple point was that we have wondered so far away from competition and capitalist concepts such as price discovery that to me, to introduce them, as the foundation of an argument, is false.

Posted by LincolnArmy on 10-18-12 04:00 PM:

Would stock prices or bond prices be at these levels without the intervention of the Fed?

For example, for the majority of Americans (or at least a large minority), the rate on the 30yr bond has no relevance to their lives. They probably couldn't get a mortgage, and if they could it would be nowhere near the 3-4% that the 30yr trades at. More likely they borrow at loan shark rates 20-30%.

Pure price discovery for me is about the interaction of buyers and sellers without the government stepping in and manipulating prices.

One other point that I suspect some may find more contentious.
Let's say that HFTs do 70-80% of the volume in may contracts. Here we have markets that are dominated by computer programmers and to expect them to give their full attention to economic fundamentals is unrealistic. Massive generalization here, but these are people whose first interest was not in the markets anyway. They are more interested in slipping and sliding or whatever the latest craze is. Their interest in finance probably doesn't spread much beyond doing an Econ 101 class. Therefore they are biased in their outlook and will take on board ideas such as the stock market always goes up in the long run as does the housing market.
Nothing necessarily wrong with that -after all markets are about differences in opinions. However when one class of traders is allowed to completely dominate trading then the market is pushed towards their bias and thus there is far more potential for the market to move away from fundamentals. (Going off at a tangent, I find this particularly the case when it appears to me that the vast majority of Americans, programmers, MBAs included, simply have not been taught to think for themselves -look how most of the public are currently caught up in this Reps Vs Dems charade)

Look at what happened in Oz. The market steams through all time highs simply because some HFT sniffed out some stops. This really didn't reflect market fundamentals and was not price discovery in the capitalistic model as I understand it.

I don't agree that no pure price discovery results in no trading or extremely smooth trading, but certainly the argument can be made that it generally results in more chaotic and volatile price movements at times.

Posted by CT10Gov on 10-18-12 04:03 PM:

This has nothing to do with what I was talking about. "Pure price discovery for me is about the interaction of buyers and sellers without the government stepping in and manipulating prices" is purely a definition made up by you, for you. It's has little relation with what everyone else understood 'price discovery' to mean: the incorporation of new information into the pricing of a security.

It's beginning to seem like you just want to rant about the government. Fine - I hate the state of the government too. We are just no longer talking about HFT or the market place.

Quote from LincolnArmy:

Would stock prices or bond prices be at these levels without the intervention of the Fed?

Posted by LincolnArmy on 10-18-12 04:19 PM:

Well my first argument was that in the case of bonds we are discovering a yield that doesn't present any useful value in the real world. So we are discovering a value that has little use, and is not a true value.

However my second part was an argument about how with the current status quo, HFT is further distorting price discovery given the actions of the Fed.

I will refrain from ranting against the govt further. My only real point is this. So often I hear complaints answered with -suck it up, this is capitalism. However the truth is that we have very little that approximates capitalism left in the US. Many pro HFT arguments are made on that basis with the jokes thrown in about abacuses, horses and cariages, you get the picture. I think that is a false premise to start from.

Posted by CT10Gov on 10-18-12 05:59 PM:

I disagree with both points. In the first instance, the sole existence of a government bond market is that... governments issue bonds. It doesn't exist without government 'intervention'. There's no value to discover where the market is willing to lend to government? Tell that to Italy or Spain.

For your second point: "HFT is further distorting price discovery given the actions of the Fed" is a bit of non sequitur. Are you saying that HFT is ultimately affecting where securities are priced at time scale greater than a few hours? And what does Fed's bond purchasing have to do with HFT, exactly?

There's another guy who argues that Central Banks directly affected HFT; Yet, he fail to identify any plausible channel of transmission, except 'leverage' (of course, HFT is not capital intensive).

Quote from LincolnArmy:

Well my first argument was that in the case of bonds we are discovering a yield that doesn't present any useful value in the real world. So we are discovering a value that has little use, and is not a true value.

However my second part was an argument about how with the current status quo, HFT is further distorting price discovery given the actions of the Fed.

Posted by LincolnArmy on 10-18-12 07:29 PM:

'given the actions of the Fed' -as in - taking the actions of the Fed as a given. I did not express myself clearly, was making no link between the two and can see how that sentence can be interpreted as you have.

With the HFTs dominating the volume is many markets, I don't think it's possible that their action wouldn't influence the direction of the markets. I have heard touted as conventional wisdom these days that 'the market still goes where it is going to, it just gets there in a different fashion'. I disagree with this.
Let any group/class of traders dominate a market and it will end up at a different price due to the inherent bias of that group. This is not a criticism of HFT, it's a criticism of any group that dominates the market.
Here's an example. On the day of the flash crash, do you think that the market would have settled at the same price that it did, if the flash crash had not occurred? I don't think it would have settled at the same price that day or even the next week.

I think that increasingly we are drifting away from real value and toward a value based on the assumptions and biases of the models used by the HFTers. Increased participation by other market participants with their own biases would help balance out those of the HFT crowd. I've seen markets dominated by certain classes of traders over my career, but never to this extent, that's why I'm making this point.

Posted by CT10Gov on 10-18-12 07:39 PM:

Here's the part where I disagree - I don't believe HFT has any persistent effect on value; Here's my argument:

So company A has a 'real value'. We don't know what it is: a momentum investor sees its value differently from a value investor, who sees it differently from a market maker. The idea is that when all these guys come together and put their money behind their opinions, and the hence the information discovery ability of the market.

So to this we now add HFT. HFT guys have no real opinion to express on the long-run value of company A. By definition, they are doing whatever it is they are doing on a tiny time scale. If they do distort the value of company A over a longer time-horizon, couldn't a value investor come in and make money on that distortion?

So, why can't HFT's distortion of real value be arbed away like any other distortion from any other group of players? Do we, in fact, have any evidence of persistent distortion of prices that persist over time?

(btw, I want to thank you for at least a sane and civilized conversation; this topic usually draws out the crazies;)

Quote from LincolnArmy:

I think that increasingly we are drifting away from real value and toward a value based on the assumptions and biases of the models used by the HFTers. Increased participation by other market participants with their own biases would help balance out those of the HFT crowd. I've seen markets dominated by certain classes of traders over my career, but never to this extent, that's why I'm making this point.

Posted by Random.Capital on 10-18-12 08:03 PM:

Quote from LincolnArmy:

Look at what happened in Oz. The market steams through all time highs simply because some HFT sniffed out some stops.

That's not at all what happened.

All times are GMT. The time now is 09:48 AM.