Learning to program - where to start? | Forex Factory

Options

Permalinks
Print Thread

Similar Threads

How to start learning Technical Analysis 29 replies

Plz help, where to start learning MQL4 to create custom indicator 18 replies

Learning to Program MQL4/5 14 replies

Learning at the speed of snail [Newbie Learning Journal] 9 replies

Learning how to program 2 replies

Learning to program - where to start?

  • #1
  • First Post: Dec 20, 2010 2:56am Dec 20, 2010 2:56am
I'm spending hours on analysing strategies and systems manually and on spreadsheets. I really need to learn to program to accurately backtest and improve my strategies. I would love to be able to program my own EA's etc.

I programmed a bit in school about 15 years ago so I do remember some basics, I'll be able to figure out the logic and use or variables etc.

My question is , just where do you suggest I start? I don't want to waste hours.
  • #2
  • Dec 20, 2010 3:14am Dec 20, 2010 3:14am
Quoting pasopvirpot
I'm spending hours on analysing strategies and systems manually and on spreadsheets. I really need to learn to program to accurately backtest and improve my strategies. I would love to be able to program my own EA's etc.

I programmed a bit in school about 15 years ago so I do remember some basics, I'll be able to figure out the logic and use or variables etc.

My question is , just where do you suggest I start? I don't want to waste hours.
Jy kan hier probeer: http://articles.mql4.com/404
Never Ever Give Up!
  • #3
  • Dec 20, 2010 3:20am Dec 20, 2010 3:20am
Quoting Cobra
Jy kan hier probeer: http://articles.mql4.com/404
HaHa Baie dankie!
  • #4
  • Dec 20, 2010 4:31am Dec 20, 2010 4:31am
I started by learning how to send orders, and started to develop my skills and ideas around that. I would look for the most basic EA that does that, and begin by studying that.
  • #5
  • Dec 20, 2010 5:01am Dec 20, 2010 5:01am
Anyway this is the basic and good rules step-by-step http://book.mql4.com/. You will see the picture about mql at all and it will be easy to improve your knowledges after. I do the same way from the same stage
  • #6
  • Dec 24, 2010 12:02am Dec 24, 2010 12:02am
you don't want to waste hours? LOL. Good one.

This is a russian programing software. You will waste more than hours. But I can save you countless hours by telling you don't waste time chasing a backtest. They don't prove anything.

Use a backtest to see that the code is working, doing what it should do etc. That's as far as you should use a backtest. Don't risk any money on a backtest only result.

The problem comes into play when you have ideas but don't know how to translate them into language the computer can understand.

I would advise you to read other people's code. Do this everyday and when you read it try to understand what the computer is doing when running it.

If you have to take the code and play with it. Figure out what it does before you move onto the next piece of code.

Alot of the functions and uses are already written so in those instances you can just copy-paste and reuse them. But you need to understand what they do and why.

You can think of code in terms of modules. Once you build a module you can reuse it over and over again.

Here is a list of common modules:

MM or money management
OrderTracking or Order Counting
GlobalVariables -- recorded information
OrderEntry
OrderClosing
TrailingStops

A module is a code set contained with an argument and opening and closing brackets that the code contained with in performs a specific function.

Inserted Code
 
// Order entry module example
 
if (Ordertotal() == 0)
    {
        if (fivesma0 > thirtysma3+mybuffer*myPoints)
             {
               // Enter Buy Long code here 
              }
        if (fivesma0 < thirtysma3-mybuffer*myPoints)
             {
              // Enter Sell Short code here
             }
     }

Once you get to the place where you are comfortable using copy paste and compiling scripts then progress further by writing your own lines of code. The more you read and play around with code the faster you will learn.

But if you think you will knock this out in 15 min or even less than 1000 hours guess again. If it was easy there would be no challenge to it and you wouldn't be here.

The only easy way to riches is inheritance or you're born into wealth... or you just happen to play the lotto at just the right time with just the right numbers. You're only limited by your own effort.
  • #7
  • Dec 24, 2010 2:05am Dec 24, 2010 2:05am
Quoting pip_seeker
you don't want to waste hours? LOL. Good one.

This is a russian programing software. You will waste more than hours. But I can save you countless hours by telling you don't waste time chasing a backtest. They don't prove anything.

Use a backtest to see that the code is working, doing what it should do etc. That's as far as you should use a backtest. Don't risk any money on a backtest only result.

The problem comes into play when you have ideas but don't know how to translate them into language the computer can understand.

I would...
Totally agree. But anyway you need a basic knowledges.
  • #8
  • Dec 24, 2010 9:40am Dec 24, 2010 9:40am
Quoting pip_seeker
Use a backtest to see that the code is working, doing what it should do etc. That's as far as you should use a backtest. Don't risk any money on a backtest only result.
Thanks a lot for the advice pip_seeker.

Why can't a backtest be used to test the profitability of a certain system?
  • #9
  • Dec 25, 2010 10:33am Dec 25, 2010 10:33am
Quoting pasopvirpot
Thanks a lot for the advice pip_seeker.

Why can't a backtest be used to test the profitability of a certain system?

FX is zero sum. A buy has an opposing sell, a sell has an opposing buy... and your broker controls whether you get in, out if you are in and stopped out or margin called.

A backtest is only a general idea. Not actual. You start throwing in slippage, missed trades etc and you will touch on the tip of the iceberg why backtesting is a movie named "3:10 to disappointment".

You cannot make something work that was never designed to work in the first place. While we are on the subject there are differences even from a demo feed to a live feed. These differences translate into real dollars on a live account. You don't win by coincidence, but you better believe you can lose by coincidence.

Part of the problem is markets always changing. Some will argue "the markets don't change - they can only go up, down, sideways." While this is true it is only one part of the puzzle.
  • #10
  • Aug 27, 2013 7:40pm Aug 27, 2013 7:40pm
I found some free tutorials about MQL4 programing. It could be useful for this thread


watch 32 videos
http://forexscalping101.blogspot.com...d-editing.html
  • #11
  • Last Post: Aug 29, 2013 11:18pm Aug 29, 2013 11:18pm
Quoting pip_seeker
{quote} FX is zero sum. A buy has an opposing sell, a sell has an opposing buy...
LOL. Tell me one market where it is different!

But in fact, backtesting can only give you an idea if something is profitable or not, never a proof. But there are ways to figure out if it is more likely that a system will be working in the future or not. Also there are a lot of pitfalls too.
There is no substitute for forward testing.
0 traders viewing now