Exchange arbitrage, is it worth digging into? - page 9
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
For "malina", the server needs to be located in the exchange's data centre and must be connected directly, not through a broker's server. And arbitrage strategies should be tested not on bars, but at least on real ticks. Also take into account delays.
I am testing on tick history.
What are the prices there approximately - colocation + direct connection? Yes, but you can start with a broker to see. There should be divergences on volatility
Try it - it won't be superfluous.
There are specific discrepancies, but very rarely.
But in the end is it still executed or not?
After all, the order only goes to the MT position after receiving a response from the exchange server about the execution of the order.
I had a buy limit on price spikes for about ten seconds, and it was hanging way above the ASC)). I could not believe how it could be)))
In the morning, they say, the exchange only at the start of trading in a rare failure and formed a queue of orders. Well, and the answers, apparently, do not have time to hand out.
Metakvot say their server can handle about one hullabaloo of bids per second. And the exchange is bragging about their servers, too.
Maybe someone is behind our backs to rip us off.) Hmm. It's only a guessing game.
Eventually it did, with a delay of up to 3 seconds.
I think it is the exchange server that slows down, not so many people trade through MT5 to slow down.
As a result, it is executed, with a delay of up to 3 seconds.
I think the exchange server is slowing down, there are not so many people trading on MT5 to slow down so much.
Checked it out, there was no slack on the direct connection today at 14 05 but there was on MT5. Sending a note of no confidence in the metaquote.
And how did you determine this? Maybe your CPU was busy with other programs and didn't have free resources? Maybe, your network was slow?
Yes, the processor purposely chooses volatility spikes to load all cores 100% and the png internet provider sets it up.
It's unrealistically hard to prove anything 100% here, so I'm not claiming anything or making accusations - just sending a note of disbelief.
In the near future I will connect to the exchange directly and, of course, do not forget to synchronize the whole thing with my favorite, the world's best terminal MT5, which does not like synchronization below a second.
If I do that I'll have more hard proof - but it's not 100% real, to have a 100 EA I need to thoroughly monitor the server part, which is unrealistic.
But, those who need it will understand.
Yes, the processor purposely chooses volatility spikes to load all cores 100% and the internet provider with png sets it up.
It's unrealistically hard to prove anything 100% here, so I'm not claiming anything or making accusations - just sending a note of disbelief.
In the near future I will connect to the exchange directly and, of course, do not forget to synchronize the whole thing with my favorite, the best in the world MT5 terminal, which does not like synchronization below a second.
If I do that I'll have more hard proof - but it's not 100% real, to have a 100 EA I need a thorough scan of server software, which is unrealistic.
But, he who needs it, will understand.
It is actually very easy to check who is lagging.
You need to ask the broker and the exchange for a specific execution time for a particular ticket.
Exchange arbitrage. Is there any point in digging?
There is no point in digging. Everything has already been dug up. )) The same goes for HFT.
But if you can't, but really want to - you can.(c) If you come up with something of your own.
Let's say HFT - you can't get there - >60% of trades in the market already. But HFT itself generates some regular movements on which you can work.