Spread between two futures - page 2

 
iiivasyaiii:

Prostatrader, good afternoon, can you tell me please, I didn't quite catch your code:

1) Did you use CopyTicksRange or CopyTicks function?

2) The description of these functions says that they can copy no more than 2000 ticks, it means that this indicator cannot analyze more distant history?

CopyTicksRange() copies as many as necessary (from - to)

 
iiivasyaiii:

Guys, I'm not arguing that bids are more accurate (although when I had a short time indicator showing three spreads: flipper, ask-bid and bid-ask, flipper most of the time went between the latter two, which suggests that on liquid instruments the flipper spread is generally adequate).

Let's then figure out how to do the asc/bid synchronization in the least resource intensive way possible and preferably with a deep history.

Here's an interesting option suggested by simplytrader (above).

There is a good example on this subject, not related to the exchange, but well illustrating the problem. If you want to do arbitrage on daily candlesticks, compare closing prices, the error will be negligible. The deeper you go down in the timeframe, the higher the error will be due to the non-synchrony of the event.

 
Dmi3:

....If you want to do arbitrage on daily candles...

Hilarious! :)

 
prostotrader:

Hilarious! :)

Well, that's how theorists like it. They take some pork futures from 1960 and arbitrage them to grain futures. They do it on a daily basis.

And they find profits that we couldn't even dream of!
 
iiivasyaiii:

Guys, I'm not arguing that the bid-ask spread is more accurate (although when I had a short-lived indicator that showed three spreads: flipper, asc-bid and bid-ask, flipper went between the latter two most of the time, which suggests that on liquid instruments the flipper spread is generally adequate).

Nothing is going to be adequate!

Why do you need the story? And to look at the spread of the selected instruments.

Those deals that were made on the instruments don't show the spread at all,

because at any timeframe the time of deals for different instruments will be different.

You (without bid and ask) don't know what the spread was at any given millisecond, and it could have satisfied

Your intentions to make an arbitrage trade.

Ideally, to view the history, you should take a tick of the first instrument and look for a correspondence of time (msec) of this tick to the time of the other

instrument. It's extremely difficult to write such an indicator, because there is a problem of showing (there are many times less bars).

You need a millisecond timeframe.

Added

But today there are 2 ways to solve the problem.

1. take bar time (MINUTES ONLY) and look for ticks in both instruments with this time (while I do so)

2. Write slices by milliseconds to a file and output the chart in your program.

 
prostotrader:

Nothing will be adequate!

Why do you need the history? And to look at the spread of the selected instruments.

Those deals that were made on the instruments do not show the spread at all,

because at any timeframe the time of deals for different instruments will be different.

You (without bid and ask) don't know what the spread was at any given millisecond, and it could have satisfied

Your intentions to make an arbitrage trade.

Ideally, to view the history, you should take a tick of the first instrument and look for a correspondence of time (msec) of this tick to the time of the other

instrument. It's extremely difficult to write such an indicator, because there is a problem of showing (there are many times less bars).

Moreover, without an analysis of the current bid and ask, you will not be able to determine whether the current conditions satisfy your intention to enter (or exit) a trade. And it would be completely wrong to analyse the flipper prices in history and the current ask/bid prices, wouldn't it?

 
Dmi3:

Moreover, without analysing the current bid and ask, you cannot determine whether current conditions satisfy your intention to enter (or exit) a trade. And it would be completely wrong to analyze the flipper prices in history and the current ask/bid prices, wouldn't it?

History is needed to determine the boundaries of spread divergence/decline.

And the analysis of current ask/bid pairs is necessary for entry/exit.

Here is the GOLD-9.20 and GOLD-12.20 spread


From the chart you can see that it was possible to enter the market either when the spread widened to 92 points or

at narrowing of spread up to 15 pips.

 
prostotrader:

History is needed to determine the boundaries of spread divergence/decline.

And analysis of the current ask-ask pairs is needed for entry/exit.

Here is the GOLD-9.20 and GOLD-12.20 spread


From the chart you can see that it was possible to enter the market either when the spread widened to 92 points or

at narrowing of spread up to 15 points.

You can see from this chart that the divergence found at the time of market opening, i.e. 10.00.00.000-10.00.01 and after the evening clearing, i.e. 19.00.00.000-19.00.01.

Neither of them allows you to enter at the clearing price. You understand that very well. So you are again a "spherical horse in a vacuum".

 
Dmi3:

From this chart you can see that the divergences found are at the time the market opens, i.e. 10.00.00.000-10.00.01 and after the evening clearing, i.e. 19.00.00.000-19.00.01.

Neither of them allows you to enter at the clearing price. You understand that very well. So it's "a spherical horse in a vacuum" all over again.

I have been trading on FORTS using arbitrage strategies since 2013 and I enter and exit the market somehow.

At the beginning of the year I opened another EIS:


 
prostotrader:

I have been trading arbitrage strategies on the FORTS since 2013 and I enter and exit the market naturally.

At the beginning of the year I opened another IIM:


Good result, congratulations.

Do you know how to enter an arbitrage trade at 10.00.00?