Jump to content
House Price Crash Forum
wonderpup

The Intelligent Allocation Of Capital-

Recommended Posts

The Banker's great claim to social value is their role in optimising capital allocation due to their ability to assess the merits of those seeking it- as a result those enterprises that were deemed stronger would thrive while the less well run or weaker ones would die.

All of which sounds great.

What should really impress though is just how fast this process of assessing the merits of cadidates for capital has become. So fast in fact that a six millisecond latency advantage in the system is now deemed so important that it's worth laying a 300 million dollar cable across the Atlantic to achive.

This really impresses me because it takes about 100 milliseconds to blink an eye. :lol:

So what- exactly- is the social value of a trading system in which shares change hands in intervals of 6 milliseconds? Is there any point at all now to the antics of these gambling addicted morons?

At what point do we say enough?

The $300m cable that will save traders milliseconds

In the high-speed world of automated financial trading, milliseconds matter. So much so, in fact, that a saving of just six milliseconds in transmission time is all that is required to justify the laying of the first transatlantic communications cable for 10 years at a cost of more than $300m.

http://www.telegraph.co.uk/technology/news/8753784/The-300m-cable-that-will-save-traders-milliseconds.html

Share this post


Link to post
Share on other sites

You might like to watch this.

I'll also add High Frequency Trading is like a cancer in the body.

Cancer consumes 8x more energy than a normal cell and when you consider these HFT algo's test the price point ie the max price you will buy at or min price you sell at, these HFT algo's deprive you of your very last penny in a share deal.

What this means is that you and countless others who deal in shares up and downthe country have less money to spend in the wider economy depriving the vast majority of businesses of which the majority are small businesses who happen to be the largest employing group in the country and thus more money goes to the traders and less to the wider economy.

But it doesnt stop there, becuase small businesses also pay the most tax either through PAYE employment or Corp tax, becuase the big companies can employ accountants for tax avoidance measures and get this, the banks and traders get to pay even less tax than a normal business would normally pay.

So not only are they taking the very last penny in a share deal, they also pay less tax to the country depriving people & this country effectively robbing society twice!

1690s, from Fr. algorithme, refashioned (under mistaken connection with Gk. arithmos "number") from O.Fr. algorisme "the Arabic numeral system" (13c.), from M.L. algorismus, a mangled transliteration of Arabic al-Khwarizmi "native of Khwarazm," surname of the mathematician whose works introduced sophisticated mathematics to the West

Sophismata are "ambiguous, puzzling or simply difficult sentences" that were used by Medieval logicians for educational purposes and for disputation about logic. Sophismata were written in Latin, and for many of them the meaning is lost

Mohamed Atta - WTC

Moor links for the lads

Some of the early scholars believed Khwarezm to be what ancient Avestic texts refer to as Airyanem Vaejah ("Ariyaneh Waeje"; later Middle Persian Iran vij). These sources claim that Old Urgench, which was the capital of ancient Khwarezm for many years, was actually Ourva, the eighth land of Ahura Mazda mentioned in the Pahlavi text of Vendidad.

Michael Witzel, a researcher in early Indo-European history, believes that Airyanem Vaejah was located in what is now Afghanistan, the northern areas of which were a part of ancient Khwarezm and Greater Khorasan.

Persian Origin > after Mongol Invasion Turkic (Sufi-isticated)

Logical Mr Spock - "the northern areas of which were a part of ancient Khwarezm"

Edited by erranta

Share this post


Link to post
Share on other sites

Milliseconds are so 2008....

The rise of the picosecond

Michelle Price, Financial News

Just when you thought high-speed cash equities trading could not get any faster, trading geeks have thrown a new concept into the mix: the picosecond.

A second is a long time in cash equities trading. Four or five years ago, trading firms started to talk of trading speeds in terms of milliseconds.

A millisecond is one thousandth of a second or, put another way, 200 times faster than the average speed of thought. In the time it took your brain to tell your hand to click on this article, a broker or market-making firm trading in milliseconds could fill hundreds of orders on an exchange.

Milliseconds, however, are now ancient history. In the past two or three years, trading speeds have been shaved down to inconceivably tiny increments: from milliseconds to microseconds, and more recently to nanoseconds.

But in recent weeks trading geeks have started to talk about picoseconds in what is a truly mind-boggling concept: a picosecond is one trillionth of a second. Put another way, a picosecond is to one second what one second is to 31,700 years.

Speaking at a London conference on Tuesday, Donal Byrne, chief executive of Corvil, a high-speed trading technology company, caused a ripple of audible incredulity throughout the room when he suggested that trading speeds could be reduced to picoseconds in the not too distant future.

For those whose brains have not instantly combusted at the concept, the rise of the picosecond prompts an obvious question: why on Earth (which spins at a rate of 460 meters a second) is it necessary to trade so fast?

The answer is simple. Firms that trade super fast effectively put themselves at the front of the trading queue and have priority over other orders. This position gives them better information on the trading behaviour of other investors and allows them to react faster.

In a bull market, speed can ultimately make for beefy profits, but in a bear market these tiny fractions matter more than ever. The potential value of millisecond, or indeed a picosecond, was vividly demonstrated during a particularly bloody period on Black Friday, October 10, 2008, when the UK market plummeted at a hair-raising £250m a second.

Link to article >

Share this post


Link to post
Share on other sites

So what- exactly- is the social value of a trading system in which shares change hands in intervals of 6 milliseconds? Is there any point at all now to the antics of these gambling addicted morons?

The theoretical benefit of HFT, I think, has to be that it helps perpetuate the belief that the market price is continuous - rather than discrete... that assets have a known value all-the-time, not just at the instant they change hands for a given price. If markets are less liquid, sinister problems become more apparent... and confidence in the religion of the 'current market price' begins to wane. As this happens, the perception of risk increases - by any conventional measure - and, with that, margin requirements increase.. and, with that, systemic leverage falls - and, overall, prices fall.

Sure, many of the trades may be speculative - but they do ensure convertibility of instruments... and that defines a functional market.

I think a far better question is this: Why is HFT profitable? If HFT is (long-term) profitable (where, in high-frequency-land, long term might be longer than a few hours) then this is evidence for an inefficient exchange. If the exchange efficiently matched prices, it would be impossible to define an HFT algorithm that yielded profit, on balance, over time. The success of HFT demonstrates the need for significantly better exchange technology. In principle, this is an opportunity for a competitor exchange that more efficiently matches trades - which would, one assumes, gradually take business from existing exchanges by offering better average prices to longer term investors.Conversely, perhaps the real problem is that hardly anyone cares as most capital belongs to someone else, anyway.

Share this post


Link to post
Share on other sites

The theoretical benefit of HFT, I think, has to be that it helps perpetuate the belief that the market price is continuous - rather than discrete... that assets have a known value all-the-time, not just at the instant they change hands for a given price. If markets are less liquid, sinister problems become more apparent... and confidence in the religion of the 'current market price' begins to wane. As this happens, the perception of risk increases - by any conventional measure - and, with that, margin requirements increase.. and, with that, systemic leverage falls - and, overall, prices fall.

Sure, many of the trades may be speculative - but they do ensure convertibility of instruments... and that defines a functional market.

I think a far better question is this: Why is HFT profitable? If HFT is (long-term) profitable (where, in high-frequency-land, long term might be longer than a few hours) then this is evidence for an inefficient exchange. If the exchange efficiently matched prices, it would be impossible to define an HFT algorithm that yielded profit, on balance, over time. The success of HFT demonstrates the need for significantly better exchange technology. In principle, this is an opportunity for a competitor exchange that more efficiently matches trades - which would, one assumes, gradually take business from existing exchanges by offering better average prices to longer term investors.Conversely, perhaps the real problem is that hardly anyone cares as most capital belongs to someone else, anyway.

Nope it is all about gunnign someone else's HFT machine plus all the rest of the market and those who "invest" in it.

Share this post


Link to post
Share on other sites

Nope it is all about gunnign someone else's HFT machine plus all the rest of the market and those who "invest" in it.

I've found no evidence yet (and I've looked) that HFT algorithms are locked in combat - betting against each other*. My strong suspicion is that HFT bets against exchanges' trade matching algorithms to the (opaque) detriment of the exchanges' other customers. HFT is doing something a bit like front-running, except that it isn't illegal as the HFT outfit are not aware of the individual trades they're front running - but can merely predict their probability distributions.

You can argue ethical or unethical - but this is to miss the point. HFT exists because Exchanges are crumby. If the exchange improves their efficiency, the HFT 'arbitrage' opportunity disappears. Conversely, if I ran an exchange, I might leave the inefficiency exploitable for a while - during which time I sell expensive contracts for high-frequency preferential access - then, once that market is no-longer profitable, I'm free to change my matching algorithm and deprive them of future profit.

--

* => I have no insider knowledge... this is just a hunch following personal research of public information.

Share this post


Link to post
Share on other sites

faster trading makes the market more liquid and more accurate to a certain degree.

behind every buy there is a sell, so the balance is equal, it makes no difference at the end of the day. high frequency trading is just arbitrage.

imagine a clock that ticks once every second, how accurate is that clock?

now imagine a clock that ticks once every half a second - you spot opportunities faster than someone else.

but at the end of the day, youre not cheating time, 1 second is still 1 second - you just have a more accurate clock.

Edited by mfp123

Share this post


Link to post
Share on other sites
faster trading makes the market more liquid and more accurate to a certain degree.

Accurate relative to what? The only way these trades can happen so fast is if the data being analysed is both immediate and easily quantifiable- which can only mean the immediate environment- which in this case is the market itself.

All that is happening here is that computers are watching other computers who in turn are watching them who in turn are watching them- there is no 'analysis' going on here- just a computer circle jerk that references nothing but itself and has already lead to the 'flash crash' event which may be the HFT version of microphone feedback as the self referencing network spirals away to infinity.

All HFT adds to the system is instability- something we are not that short of at present.

Share this post


Link to post
Share on other sites

I've found no evidence yet (and I've looked) that HFT algorithms are locked in combat - betting against each other*. My strong suspicion is that HFT bets against exchanges' trade matching algorithms to the (opaque) detriment of the exchanges' other customers. HFT is doing something a bit like front-running, except that it isn't illegal as the HFT outfit are not aware of the individual trades they're front running - but can merely predict their probability distributions.

You can argue ethical or unethical - but this is to miss the point. HFT exists because Exchanges are crumby. If the exchange improves their efficiency, the HFT 'arbitrage' opportunity disappears. Conversely, if I ran an exchange, I might leave the inefficiency exploitable for a while - during which time I sell expensive contracts for high-frequency preferential access - then, once that market is no-longer profitable, I'm free to change my matching algorithm and deprive them of future profit.

--

* => I have no insider knowledge... this is just a hunch following personal research of public information.

With 60-70% share of all equity market trades then are they not effectively mainly trading against each other?! HFT ecists beause it suits those that write the rules to have them there there are capital gains rules, there are stamp duty and other charges, there are some market trading scahemes that have zero tax, there are some that allow gearing.

Gearing and debt hasve caused enormous disruptions and damage, no reason to believe that overt frequency trading is not doing the same.

Share this post


Link to post
Share on other sites

Accurate relative to what? The only way these trades can happen so fast is if the data being analysed is both immediate and easily quantifiable- which can only mean the immediate environment- which in this case is the market itself.

All that is happening here is that computers are watching other computers who in turn are watching them who in turn are watching them- there is no 'analysis' going on here- just a computer circle jerk that references nothing but itself and has already lead to the 'flash crash' event which may be the HFT version of microphone feedback as the self referencing network spirals away to infinity.

All HFT adds to the system is instability- something we are not that short of at present.

accurate in terms of the arbitrage.

the computers are just competing against each other. if you put 2 humans against each other, the person who can spot the arbitrage first i.e see that you can buy 1 share from bob at £1 and sell to mike at £1.01, has that advantage. these days its done by a computer.

every buy is matched by a sell at the end of the day. there is nothing wrong with a flash crash - someone elses loss is matched by an opportunity for someone else.

a share that is too cheap is an opportunity for someone to buy, a share that is too expensive is an opportunity for someone to sell. the net effect of making things faster is zero.

Share this post


Link to post
Share on other sites

The Banker's great claim to social value is their role in optimising capital allocation due to their ability to assess the merits of those seeking it- as a result those enterprises that were deemed stronger would thrive while the less well run or weaker ones would die.

All of which sounds great.

What should really impress though is just how fast this process of assessing the merits of cadidates for capital has become. So fast in fact that a six millisecond latency advantage in the system is now deemed so important that it's worth laying a 300 million dollar cable across the Atlantic to achive.

This really impresses me because it takes about 100 milliseconds to blink an eye. :lol:

So what- exactly- is the social value of a trading system in which shares change hands in intervals of 6 milliseconds? Is there any point at all now to the antics of these gambling addicted morons?

At what point do we say enough?

http://www.telegraph.co.uk/technology/news/8753784/The-300m-cable-that-will-save-traders-milliseconds.html

Sounds like more tail wagging the dog, same as the way banking became an 'industry' in it's own right.

Share this post


Link to post
Share on other sites
the computers are just competing against each other. if you put 2 humans against each other, the person who can spot the arbitrage first i.e see that you can buy 1 share from bob at £1 and sell to mike at £1.01, has that advantage. these days its done by a computer.

So where in this process does the bit about intelligent allocation of capital come in?

Share this post


Link to post
Share on other sites

I think a far better question is this: Why is HFT profitable? If HFT is (long-term) profitable (where, in high-frequency-land, long term might be longer than a few hours) then this is evidence for an inefficient exchange. If the exchange efficiently matched prices, it would be impossible to define an HFT algorithm that yielded profit, on balance, over time. The success of HFT demonstrates the need for significantly better exchange technology. In principle, this is an opportunity for a competitor exchange that more efficiently matches trades - which would, one assumes, gradually take business from existing exchanges by offering better average prices to longer term investors.Conversely, perhaps the real problem is that hardly anyone cares as most capital belongs to someone else, anyway.

The answer is don't know yet. I spoke to an algorithm programmer (not HFT) and he says if you have a statistical advantage of say 2%, it will take several years to confirm that the advantage is true. He also said that investing in HFT systems is pretty high risk - if your competitor then become 1 picosecond faster, you lost all the investment.

So, even if HFT is profitable on operating basis, it might not be when overall cost are taken into account (because 3 days later someone comes up with something faster and the whole million pound computer system can then be consigned to the bin)

Share this post


Link to post
Share on other sites

Accurate relative to what? The only way these trades can happen so fast is if the data being analysed is both immediate and easily quantifiable- which can only mean the immediate environment- which in this case is the market itself.

All that is happening here is that computers are watching other computers who in turn are watching them who in turn are watching them- there is no 'analysis' going on here- just a computer circle jerk that references nothing but itself and has already lead to the 'flash crash' event which may be the HFT version of microphone feedback as the self referencing network spirals away to infinity.

All HFT adds to the system is instability- something we are not that short of at present.

there is no proof that HFT lead to the "Flash Crash", thats an assumption based on nothing, the market moved and a reason has to be found for it after the event, that is the nature of the news everyone always wants a reason when no definitive one exists which is why at any time an opposite move will generate the same story (ie when an interest rate cut is used as a reason for both the market going up and down at different times). There doesnt seem to be anything unnatural about the flash crash which is proven by the fact that rather than recovering the market was confirming the new lows a month later, was HFT responsible for the October 29 move, or the 87 move, the nikkei was extremely volatile in 1990

The volatility is simply a function of bear markets, they contain more emotion, when you look at the data of all these market moves over a century the fractal formations dont show any difference at all

And i dont see how it effects investors in the slightest, they decide something is fundamentally cheap at a price, any volatility doesnt change the fundamental reason for buying at the price they chose, i didnt see any complaints about HFT when the market was going up from 2003, it is basically trotted out when the market is falling (when someone has made a bad investment and doesnt want to blame the real culprit, their own analysis. The more trades the more liquidity the better the price and the lower the brokerage. Its basically a chimera much like shorting which is everpresent ( in bull markets and bear markets) that is used by people to justify their mistakes, theyd be far more effective trying to understand why their analysis was wrong rather than blaming the weather.

Fundamentally you cant avoid volatility in a bear market and the bigger the bear the greater the volatility anything else would be unnatural, the problem is people have no grasp of history and simply ignore the lessons of the past and assume it should be different this time, when its not there has to be blame apportioned no matter that the chosen thing to blame is irrelevant it makes us feel better having a crutch.

Thats not to say its not pointless as a way of companies gaining access to capital, but whilst we like to think thats the reason/benefit of the stock market, it has also been about speculation since it was first formed

Edited by Tamara De Lempicka

Share this post


Link to post
Share on other sites

http://www.bankofengland.co.uk/publications/speeches/2011/speech509.pdf

Financial experts at their best! :lol::lol::lol:

TBH this speech is well worth reading as its so relevant to this thread and algo trading.

are you saying the longer it takes for price discovery is better, it seems to me the longer it takes the more damage it will cause, as you can see by the south Sea and even your own UK HP bubble

Nowheredo i mention they are financial experts they are speculators, it is human nature to bet on two drops of water running down the window, thats unlikely to change

And i repeat there is no proof the flash crash wouldnt have happened anyway, if it was unnatural it wouldnt have confirmed the lows a month later, you are highlighting what i said an opinion based on nothing we remain unsure quite what caused the Flash Crash or whether it could recur that is the important bit the same as the 87 and 29 crash, human emotion and an exhaustion of buyers caused the move and it will continue to do so as long as the stock market exists

Edited by Tamara De Lempicka

Share this post


Link to post
Share on other sites

are you saying the longer it takes for price discovery is better, it seems to me the longer it takes the more damage it will cause, as you can see by the south Sea and even your own UK HP bubble

Nowheredo i mention they are financial experts they are speculators, it is human nature to bet on two drops of water running down the window, thats unlikely to change

And i repeat there is no proof the flash crash wouldnt have happened anyway, if it was unnatural it wouldnt have confirmed the lows a month later, you are highlighting what i said an opinion based on nothing we remain unsure quite what caused the Flash Crash or whether it could recur that is the important bit the same as the 87 and 29 crash, human emotion and an exhaustion of buyers caused the move and it will continue to do so as long as the stock market exists

Trading the cra out of something has nothing to do with price discovery, and as mentioned absolutely nothing to do with allocation of capital - it is a bit like comparing the funding of a car plant to make a new car with making a machine that ships the paint of passing cars on the road. Look at the mortgage/housing market - through the bubble certification and mortgage approval times dropped - same day mortgage approvals, no time to check docs / earnings - not to worry just make loans with no docs available. In the case of the mortgage market it just left all concerned with no time to really think about the consequences, caught up inthe bubble they just went with it.

Share this post


Link to post
Share on other sites

The answer is don't know yet. I spoke to an algorithm programmer (not HFT) and he says if you have a statistical advantage of say 2%, it will take several years to confirm that the advantage is true. He also said that investing in HFT systems is pretty high risk - if your competitor then become 1 picosecond faster, you lost all the investment.

So, even if HFT is profitable on operating basis, it might not be when overall cost are taken into account (because 3 days later someone comes up with something faster and the whole million pound computer system can then be consigned to the bin)

That HFT is high risk I absolutely agree. I also agree that even the smallest delays can lead to the difference between having an advantage and having no advantage.

I think the idea of it taking years to identify if you have an advantage is a grey area, at best. If this were an online game, rather than real money, and I could execute every second, I'm pretty sure that I could tell you with stunning accuracy, after an hour, what advantage I had during that hour. The problem is that this gives me next to no information about whether or not I'll have the advantage during the next hour. If the opportunity to test only arises infrequently, it could take arbitrarily long to identify if a strategy gives any advantage at all... and if you want to be cautious about risking capital - you're left with a humdinger of a problem... none of the conventional measures of risk really help... you're on-your-own trying to estimate the likelihood that what's important about the pattern of other's bids and offers remains true while you trade it... and therein lies a big problem. An advantage would be had by anyone who knows how the exchange works internally... similarly, there are advantages to having the most accurate tick-data... and there are real advantages if you can analyse strategies over histories of tick data. The final killer problem is that, for small yields, realistically, you need big stakes - in order to overcome fixed costs.

I'm not as negative as to consign the hardware to the bin, however... I think that there are myrriad opportunities to systematically trade, as long as prices are volatile - which is, pretty much, a given in an open market. "All" that you need is the right approach...relying solely on speed to exploit deficiencies of the exchange is one strategy - and it will work for, at least, one participant - until the exchange improves its procedures and the opportunity vanishes. HFT will make money until it doesn't - and that's about all you can say about it. The gamble by the exchanges is that they will make more from HFT speculators throwing good money after bad than they will from short-term losses. The real winners will be the insiders who know, from information that should be considered privileged at the exchange, when the opportunity will turn sour - who can then sell their HFT business shortly before the HFT opportunity expires. The real losers will be the buyers of past-their-sell-by HFT operations.

Share this post


Link to post
Share on other sites

imagine a clock that ticks once every second, how accurate is that clock?

now imagine a clock that ticks once every half a second - you spot opportunities faster than someone else.

now imagine the banker's 5000 quid Swiss watch that beats.... 8 times a second ! (although it is less accurate than the 10 euro quartz watch worn by his customer and bought from Lidl). Still it is better than a Swiss Rail clock which, for some bizarre reason, beats 0.98 times a second so it never tells the correct time.

Erm, I digress.

Share this post


Link to post
Share on other sites

No its not. I can put in a sell order for something above the market price and it will stay unsold, likewise I can put an order in to buy at a ridiculously low price and I will have nothing to show for it.

...And if you do enough of these small orders fast enough you can discover the top price someone is willing to pay and then sell at that.

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...

  • Recently Browsing   0 members

    No registered users viewing this page.

  • 338 Brexit, House prices and Summer 2020

    1. 1. Including the effects Brexit, where do you think average UK house prices will be relative to now in June 2020?


      • down 5% +
      • down 2.5%
      • Even
      • up 2.5%
      • up 5%



×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.