Bid && Ask && Spread - page 8
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
But nevertheless there is such a concept. And it is also present in the link you cited. So you can't forget it.)))
...
They will be different. This is because for this strategy only two prices are very important - High Bid and Low Ask. The tester's data on High Bid will be accurate from the history. But Low Ask - no, because it will be modeled = Low Bid + Spread.
I will try to test it sometime. Thanks.
What percentage of traders do you think understand that a narrow spread is not an indicator of favorable trading conditions (all other things being equal)?
To be honest, I don't know. But I know that this percentage is formed from the number of those traders who have done enough research in this area. )))
For myself, or more precisely for my strategies at this point in time, I tend to believe that narrow spreads are not an indicator of a profitable trading strategy. Profitability of trading conditions and thus the profitability of a trading strategy depending on the spread size becomes more apparent the smaller is the TF. But the trading strategy (algorithm) must be profitable. ))) I have not developed a strategy for TF less than H1 yet. But I am interested in it.
And in other terminals you cannot implement your idea at the moment?
Are other terminals failing to implement your idea at the moment?
What kind of idea?
More specifically, is it possible to test any of your strategies more accurately in other terminals? Am I correct in assuming that the real problem appears in those strategies that are set up on small TFs? Simply, if there is such a dependency on Ask/Bid price positions, then I think such strategies should be abandoned. Probably there are many trading strategies which simply do not work in real conditions because of such dependence. You need a tick history for a long period, which is unlikely to happen. If we are given, for example, a tick history for a small period, we will not know how much the difference between Ask and Bid will change at the moment we launch the Expert Advisor in the trade. The conditions may change not in our favor. So, here we will already have the same song, like: "The market has changed" in case of a floating spread, but already at the level of the Ask/Bid difference. If the spread is more or less stable, then why is there such a difference in tests? And if setting pending orders closer strongly changes the result, then again, it is worth considering whether such a TS is worth using at all.
Actually maybe there are a lot of options that I just don't see at the moment. I just don't see what you see. :)
I am interested in scalpers in prospect, but just as another type of strategy in my strategy portfolio, and it is questionable, because I have not made any serious tests yet. Good luck.
There is its own calculator (tester), and other terminals (including Metatrader) are only used as a trading API. There is a separate branch for proper understanding of the subtleties of writing your own calculators.
You have a very poor knowledge of the total liquidity of strategies based on "noise". In FOREX you can get more than $1 mio profit per month from one FI on "noise" according to rough estimates. And this is not theory, but practice.
And for many noise strategies there is no need in tick history, the M1 OHLC Bid + OHLC Ask will be enough. Again, this is also from practice.
Generally, assessment of the riskiness of a TS strongly depends on the quantity and quality of trades. The "shallower" the work, the more stable is the used pattern - the market inefficiency.
TCs can be (roughly and conventionally) divided into the following categories (FOREX) according to the increase of risk:
P.S. A post concerning arbitrageurs and noisy TS:
Есть интересные способы реализации HFT - это когда место будущей неэффективности предсказывается. Тогда имеется возможность заранее (хотя бы за секунду) отправить на соответствующий ценовой уровень лимитник. Но там свои нюансы тоже, однако, вопросы latency не стоят так остро, как в классической схеме.
That is, you can test it with your own implementations. What's the problem then? If you have many variants that work, use those that work. Maybe I'm missing something that you're missing...
You have very little knowledge of the total liquidity of strategies based on "noise".
More accurately, you have none at all. Haven't studied the subject of "noise" yet.
In FOREX you can get more than $1 mio profit a month from one FI on "noise" according to rough estimates. This is not theory but practice.
Where can you read the results of the practice? Whose practice? What risks does this million come with? 1 mio from 100 thousand? 1 mio from 1 mio? 1 mio from 10 mio? It is customary to give figures as a percentage per annum at all.
TC on the increase of risk can be (roughly and conditionally) divided into the following categories (FOREX):
I'm stuck in the third category so far, then. )))
That is, you can test with your own implementations. What is the problem then? If you have many options that work, then use the ones that work. Maybe I'm missing something that you're missing...
I use and have no problems, nor do I have any complaints about Metaquotes. Raised the topic of improving the current MT5 tester and in general some departure from outdated ideas about many stages of TS creation. I want to improve the literacy of all market participants, from traders, brokers to trading platform developers. Now there is such an opportunity.
Where can you read the results of the practice? Whose practice? With what risks does this million come? 1 mio from 100 thousand? 1 mio from 1 mio? 1 mio from 10 mio? It is customary to give figures as a percentage per annum at all.
There is a notion of a ceiling on market inefficiency being used. Profits cannot grow exponentially, from a certain stage they grow linearly and the profitability of a strategy is measured in absolute values of the relevant ceiling.
This kind of practice can only be personal.
...... Basically the maximum spread will give the high asc, and the average is not clear at all. So how do you work with levels? After all, when you sell, the stops will be triggered by the asc. And to understand where the asc is in the past, we need the maximum spread.
The maximum spread can be anywhere on the bar. And anyway, think again.
For five-minute bars we should specify the maximal one from the minutes?
And for 1-hour bars it will not be crazy with this logic?
The maximum spread can be anywhere on the bar. And in general, think again.
For five-minute bars, should we specify the maximum of the minute bars?
And for the hour bars? Wouldn't this logic be crazy?