Pair trading and multicurrency arbitrage. The showdown. - page 14

[Roman Poshtar]

Roman Shiredchenko #:

You don't have to jump to conclusions too soon.... ;-)

I'll throw the link here - one.

It is not the truth in the extreme instance - but arbitrage on the Moscow stock exchanges + posts left-right....

h ttps:// www.mql5.com/ru/forum/372280/page28#comment_27072133

Otherwise there are trolls here already.... :-)

Two.

h ttps:// www.mql5.com/ru/forum/448777/page12#comment_47736267

Thank you. We'll read it.

[Roman Poshtar]

I'll try to get a new owl up today, mates. In about two hours.

[mytarmailS]

mytarmailS #:

It doesn't work because there's no linear relationship between the instruments, you know?

If there is no such thing, then there is no arbitrage, and all the beautiful capital curves are the result of optimisation, which will not work on new data.

If there is no linear relationship, nothing will work, if there is a linear relationship, everything will work.

The spread on new data should look at least like this.

Ideally, it should look like this

[Roman Shiredchenko]

Roman Poshtar #:

Thank you. We'll read it.

h ttps:// www.mql5.com/ru/forum/372280/page23#comment_26218400

Basics in arbitrage trading with material like what text to google to understand - like pair trading - from this post you can read....

[Roman Poshtar]

So friends, owl by indicator ind_2_linep1. It works at M1 opening prices, testing mode is corresponding.

EURUSD-GBPUSD results for 13 years, default parameters

Parameters:

FirstEnter - first divergence according to the indicator

NextEnter - refill;

Exit on the reverse crossover, lots are the same. In order to make lots according to the indicator, I think you need to edit the code.

[mytarmailS]

Roman Poshtar #:

So friends, owl by indicator ind_2_linep1. It works on M1 opening prices, testing mode is corresponding.

EURUSD-GBPUSD results for 13 years, default parameters

Parameters:

FirstEnter - first divergence by indicator

NextEnter - refill;

Exit on the reverse crossing, lots are the same. To make lots according to the indicator you need to edit the code, I think.

It's not arbitrage.

I can prove in two steps

1. it's not arbitrage

2. It doesn't work.

[Roman Poshtar]

mytarmailS #:

it's not arbitration.

in two stages, I can prove

1. it's not arbitration

2. It doesn't work.

This is not evidence. If that's what you claim, then provide your evidence. What you did to test it, what results you got, etc.

[mytarmailS]

Roman Poshtar #:

That's not proof. If that's what you claim, then provide your evidence. What you did to test it, what results you got, etc.

Of course it's not proof, I'm just making a point.

I just have one question before I start.

what do you think the spread is?

[Roman Poshtar]

mytarmailS #:

of course not proof, I'm just making a point.

I just have one question before I start.

what do you think the spread is?

I'm using the indicator now. What about you? Share.

[mytarmailS]

Roman Poshtar #:

I'm on the indicator now. What about you? Share.

I apologise for the inaccuracy in my question, I meant what spread (commission) you put in the calculation of the profit curve....

real or constant.

I plot a linear regression between two assets (classical).