Discussion of high-frequency trading on MT5 - page 56
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
As far as I understood, hrenfx here is talking about some extension of the HFT concept, which he himself invented (maybe together with Denis).
I do not agree with this interpretation of HFT, although at the beginning it seemed to me interesting:
hrenfx: Например, синтетический арбитраж - это цели по каждому символу иногда в сотню пунктов. Длительность удержания позиции иногда часы. Но это HFT-стратегия.
If synthetic arbitrage could be considered HFT, then that would be the name for any strategy at all when trading with a broker offering favourable terms to the trader.
I stick to the more standard understanding, according to which HFT is the use of technological advantages (minimum pings and supercomputers, for example) in order to squeeze a huge number of crumbs out of "suckers". "Huge numbers" means at least tens of thousands of orders per session. And this requires a very serious amount of money, which hardly any trader has.
Maybe HFT has something to do with quants, but my impression is that quants are a business based on complex and not at all obvious market matmodels. And HFT is primitive in terms of mathematics.
Thank you. Quite valuable information leading to interesting thoughts. The only thing about the "short" position retention time, I would like to estimate the MO of such time quantitatively. 1sec, 0.1...? For FX-HFT.
I understand that the boundary is relative, but still. Pipsing is 10 min +- order (x10), when the MO of holding is 100 min, it is intraday, below 1 min it is possible FX-HFT, but since any term is a product of social group agreement, I am waiting for confirmation.
I've been doing HFT for a long time, but not on FX. Now I'm working on more capital intensive strategies. So I can't say anything about the average position holding time for FX-HFT.
For the most recent strategy on futures the 95% quantile of position holding time was less than 100 milliseconds. For the earliest strategy the average position holding time was around 10 seconds.
On the other hand, when I was doing market making on one of the dull but promising instruments (that algorithm definitely fell into the category of HFT as it was based on very fast order rebidding) - position holding time was about 10 minutes but it was hedged within 100 milliseconds after opening. >50k orders per day, fill rate <1%. Very high competition :(
What is level3 ?
It is a flow of information about each order entering the trading system that is modified or cancelled. Orders are usually impersonal, i.e. it is impossible to associate an order with any particular trader.
The state of the market window (Level2) and the trade flow can be exactly restored from this flow.
I do not know if such information is available on FX. It's not a problem to get it on stock markets (except in terms of technology...).
I worked for 2 major funds, where there were separate, casino-like financial psychology experts (game psychologists..etc.) to develop the most profitable patterns, profit/loss chains that give the maximum "involvement" of the investor, his pleasure, hopes, with minimal resulting payoff (for him). And such specialists are sometimes paid more than analysts and risk managers. Also with DCs, they employ experts in PR and financial psychology who resonate with society, their aim is to pick up and retain.
Wow, just in case I looked it up, competitors are recruiting programmers and physicists/mathematicians.
Please link to a similar vacancy (want to laugh).
I don't know what to do with them, but I'm not sure if they are right or wrong. This is a scam. Luckily nobody will listen to me, because your losses are my profits, at least partially, but mostly profits of DCs, of course.
Welcome to the exchange! No cheating, you'll have a unique opportunity to get absolutely honestly lose to robots, trading by "classic". :D
I stick to a more standard understanding, according to which HFT is using technological advantages (minimum pings and supercomputers, for example) in order to squeeze a huge amount of crumbs out of "suckers". "Huge numbers" means at least tens of thousands of orders per session. And that requires a very serious amount of money, which hardly any trader has.
The costs depend on the market on which you are going to trade. On underdeveloped markets the entry threshold is quite low.
Maybe HFT has something to do with Quantitative, but my impression is that Quantitative is a business based on complex and completely unobvious market matmodels.
The Quantitative approach is characterised by the lack of subjectivity associated with the fact that conclusions are drawn from the results obtained by calculation, rather than from"Elliot waves" and the like.
"Quants" can be roughly divided into two camps:
1. those whose research is based on theoretical considerations of "how things should be". Most often, model development is accompanied by the use of fierce mathematics.
2. those who based their research on real data - practitioners of the empirical approach. Here everything is usually not so terrible and is available for understanding to mere mortals.
And HFT is primitive in terms of mathematics.
Here I would argue :) Many widely known models are quite complex to put it mildly, but not in terms of understanding the concepts, but precisely in terms of the mathematics used.
So what are we talking about here in this thread? Of course we are talking about FOREX, and probably also about fairly liquid instruments, and not about some krona or rand.
And HFT is primitive in terms of mathematics.
I would arguehere :) Many well-known models are quite complicated to say the least, but not from the point of view of understanding concepts, but precisely from the point of view of mathematics used.
Exactly HFT models - or are they "quantum" models in general?
And yes, I was talking specifically about conceptual complexity. Clearly, the mathematics of the calculations themselves have to be somehow optimized to count faster... counting in tens of milliseconds. But it's not so much mathematics as computer science.
Exactly HFT models - or "quantum" models in general?
And yes, I was talking specifically about conceptual complexity. Clearly, the mathematics of the calculations themselves have to be somehow optimised to count faster... It counts in tens of milliseconds. But it's not so much mathematics as informatics.
It is HFT. There are gentlemen who like to write out some stochastic stochastic diffusers and Hamilton-Jacobi-Bellman equations when solving problems about optimal liquidity provision. Clearly, for practical purposes, this is greatly simplified.
But what are we talking about in this thread? Of course, about FOREX, and surely about sufficiently liquid instruments, not about some crowns or rands.
As I understand it, the topic of discussion is the application of MT5 to HFT. I think MT5 is positioned not only as a terminal for FX.
Ok, never mind :)
Wow. Nah, that's too much.
OK, forget it :)
Yeah, I agree.
This is just more theoretical nonsense about FX-HFT being impossible. What's more, the claims that FX-HFT is possible are also nonsense, if they also make theoretical sense.
No one here has tried FX-HFT in practice. I, on the other hand, have tried it, thanks to a unique and universally available darkpool, which took millions of dollars to create. I suggest you just take it and try it, as the solution is available to anyone.
For the first time, a toolkit that only banks and high-tech hedge funds could use, because the cost of creating it is huge for a physicist, not to mention the other fixed costs, has become available to the mass client.
You see FOREX from a single trader's bell, and in a highly distorted way. It so happens that I know the FOREX-industry from different sides. Somewhere very well, somewhere not enough. You would listen to the practice and just give it a try. And on the basis of the experience gained in HFT, talk about its possibilities, rather than engage in theoretical speculation "to be or not to be".
This is just more theoretical nonsense about FX-HFT being impossible. What's more, the claims that FX-HFT is possible are also nonsense, if they also make theoretical sense.
No one here has tried FX-HFT in practice. I, on the other hand, have tried it, thanks to a unique and universally available darkpool, which took millions of dollars to create. I suggest you just take it and try it, as the solution is available to anyone.
For the first time, a toolkit that only banks and high-tech hedge funds could use, because the cost of creating it is huge for a physicist, not to mention the other fixed costs, has become available to the mass client.
You see FOREX from a single trader's bell, and in a highly distorted way. It so happens that I know the FOREX-industry from different sides. Somewhere very well, somewhere not enough. You would listen to the practice and just give it a try. And on the basis of the experience you gained in HFT, talk about its possibilities, rather than engaging in "to be or not to be" theoretical speculation.
))))))))))))) I'm lying under the table ... millions of dollars - that word combination must have a great effect on others )
Don't be so hard on your ears, you unique darkpool ... )))))
I won't debate with you as there has always been enough shit in the world.
Here's a simple post. Any of the third party developers mentioned there have spent > $1 mio on their development.
Those in the know (Renat for example) know and can estimate the order of magnitude of what similar and much more complex developments (darkpools) cost. Moreover, few have any idea how such a darkpool can be created organizationally and technologically, so even if desired, it may not be possible to repeat immediately.
I am only aware of two such technologies. One is at my broker, the other is at G**X broker, recently announced (there also large sums of money and years were spent to create the technology), but it is not so perfect yet. And my broker has an understanding of what needs to be done to improve the quality of trading services. Improvements are being made all the time.
The rest of you, please note that I have not spent a penny on the mentioned darkpool. It has been created without my direct involvement.
I benefit from it on general rights, available to everyone. It is my choice, not yours. I merely informed my choice by giving my reasons for it.
I almost forgot:
Because my broker has the best prices, the highest liquidity and the best market execution, thanks to ahead of the time aggregation technology.
There is such a PR method as well.
P.S. The chatter was triggered by an interview that forexmagnates persuaded me to do. Probably shouldn't have done that.
Good confession, thank you. Although it's not the first time I've seen something similar from you.
Good thing the term "HFT" isn't there - even with the latest technology. Frankly, I'm a little perplexed when talking about technologies that require investments of millions and tens of millions of dollars.
And I have already said that it is this broker that I will be looking at closely.
It's not really about the advertising of the resource, it's about the fact that such technology has been implemented and will put pressure on the kitchens of brokers, who will be forced to move towards the trader.
I won't debate with you as there has always been enough crap in the world.
Here's a simple post. Any of the third party developers mentioned there have spent > $1 mio on their development.
Those in the know (Renat for example) know and can estimate the order of magnitude of what similar and much more complex developments (darkpools) cost. Moreover, few have any idea how such a darkpool can be created organizationally and technologically, so even if desired, it may not be possible to repeat immediately.
I am only aware of two such technologies. One is at my broker, the other is at G**X broker, recently announced (there also large sums of money and years were spent to create the technology), but it is not so perfect yet. And my broker has an understanding of what needs to be done to improve the quality of trading services. Improvements are being made all the time.
The rest of you, please note that I have not spent a penny on the mentioned darkpool. It has been created without my direct involvement.
I benefit from it on general rights, available to everyone. It is my choice, not yours. I simply informed my choice by giving my reasons for it.
I'm not going to debate and .... another 9 line bullshit )))))
Boy, how does a kitchen maid know about HFTs ? You only got into the stock market yesterday.
Metaquotes has given birth to something incomprehensible for 5 years, while others come out with full-fledged systems in 2-3 years and work with a lot of brokers on the exchange and do not whine about how hard it is.