Machine learning in trading: theory, models, practice and algo-trading - page 615
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
With Dll and a request to install mt5, R and necessary packages it is probably unreal to get there.
As I understand it: in your computer mt5, DLL to R, R client, and there, in the distance, R server for your client and execution of computationally intensive code on R, and there is a lot of it in R: crossvalidation, monte carlo, bootstrap, about different kinds of optimization I am silent.
As I understand it: you have mt5 on your computer, DLL for R, R client, while there, far away, is R server for your client and execution of computationally intensive code on R, and R is full of it: crossvalidation, monte carlo, bootstraping, I'm not mentioning different kinds of optimization.
Well, if so, then we should look into it. Thank you!
With Dll and a request to install mt5, R and the necessary packages, it is probably unreal to get there.
You can get anywhere and it is quite simple and uniform.
In the DLL we make an IP client or several IP clients. (You can also use Pipes or Memory Map Files). We make an IP server in the Eexec. We send/receive information via IP channel with text lines, e.g. CSV. Then we do what we want in the exe-server and with what we want.
As a result, MT and the executable are communicating with each other.
There's also the question of price, the author of the article himself has done the math:
You can probably buy a refurbished Z620 for 600 EUR on Ebay, which is approximately what it costs to run the c5.9xlarge instance for 17 days (on-demand).
You can get anywhere and it is quite simple and uniform.
In the DLL we make an IP client or several IP clients. (You can also use Pipes or Memory Map Files). We make an IP server in the Eexec. We send/receive information via IP channel with text lines, e.g. CSV. Then we do what we want and with what we want.
I am writing about a real dll: mt4R.
What you are suggesting is very interesting, but I want to see a real example according to your scheme.
There's also the question of price, the author of the article himself has done the math:
You can probably buy a refurbished Z620 for 600 EUR on Ebay, which is approximately what it costs to run the c5.9xlarge instance for 17 days (on-demand).
If Ebay, buy the processor, not place the calculations.
If Ebay, then buy the processor, not place calculations.
I think he meant the whole BU server, not just the processor
It's easier to buy and use, if it works, great, if not - you can always sell it for about the same price. And for the calculations will give the same or more money and already without refund.
If Ebay, then buy the processor, not place calculations.
But these are all thoughts, and practice with concrete experience of use is valuable.
To me, it's information to think about.
But I always try to stick to the general line, and today it is R, and the "general line" is determined not only by the composition and prevalence of R, but by the fact that today it is part of Microsoft.
I am writing about a real dll: mt4R.
What you suggest is very interesting, but I would like to see a real example according to your scheme.
I have this scheme really works with QuickBooks via Lua <-> DLL <-> EXE. Nothing complicated, any programmer is able to do it. I have described the scheme.
If anyone wants to do it, I can give you some advice. I don't have an MT in the pipeline.
If anyone needs it for the Quicksilver, I can send it to you. You will need a VS. For MT, you need a little remodeling of the DLL.
No one wants to do something new if the existing MT-R communication tool works.