Don't tell me then that TA doesn't work

 

Almost every month there is a thread on this forum in which another whiner, gundos or luder states that technical analysis is self-defeating.


To disprove this hypothesis once and for all, I have developed and configured a program that performs optimization on two fragments of historical data and runs a lice test on the third fragment where no optimization was performed. The probability of a successful test outside the optimization sample is over 90%. At least, no matter how many times I tried to use EURUSD, I never saw a single failed test.


As a result, it turns out that the reasons why TA "doesn't work" are:


1. The crooked hands of the poor trader.

2. Various left methods and strategies which are obviously not suitable for trading, and were created for suckers.

3. broken and broken quotes, with which the servers of some kitchens are clogged.

4. The high overhead costs for some financial instruments: spreads, swaps, broker commissions, slippages, etc., etc.

5. The desire to make money quickly, instead of stupidly not to lose it.


I.e. in capable hands TA works and is very successful. But if before it all was mostly unsupported and unsubstantiated, now there is a program, which by itself (and not by crooked hands) puts gold-dust.mq4 in the correct folder of the EA, optimizes it, tests and adjusts. All that remains is to make sure and trade.


Now, as for the program, which can be downloaded from the latest downloads page: http://gold-dust.info/ru/downloads. In the archive, there is an installation file. The distribution is free. It is not a demo version but a live one. To successfully start the program you need to have Java JRE or JDK preinstalled on your computer, which can be downloaded from http://java.com.


After that install program, read carefully, that it is informed, and also Readme.rtf file in a directory in which software has been installed. Now run it. If everything is done correctly, and not through the back door, then the success of the test outside the optimization sample is ensured.


Warning. The program was tested only under WindowsXP. No guarantees under other axes can not give, because there need to configure permissions for writing files by the program to the folder in which it was installed.

Just in case, check the file for viruses. DrWeb didn't find anything, but God protects the well.

 

Well, Richie's been called a whiner ...... ;) https://www.mql5.com/ru/forum/129868

as for the subject: could you please upload the Readme.rtf file separately, I would like to read what you have packed in the installer first

 
IgorM:


for the subject: please send the Readme.rtf file separately, I would like to read what you have packed in the installer first


Receive and sign for
Files:
readme.zip  115 kb
 
Reshetov:

Receive and sign for it

thanks, I don't want to sign on the monitor ))))
 

Under Win7 it worked without any surprises.

I wish I knew what the test sample was...

 
Reshetov:

technical analysis is self-defeating.

I agree. Here's the right one!, and the next versions improved 2,3,4,5 can be distributed for a fee (you need bread and butter!)
 
alsu:


Under Win7 it worked without any surprises.

More often than not, some software will not run on Vista. The reason is quite simple - they put an OS on their computer and don't know how to configure it.

alsu:


I wish I knew what the test sample was...


You can find out the test sample if you run the terminal and bring up the tester Ctrl + R. The settings of the last test remained there, but in order to get the EA settings, you need to click "EA Properties", then click "Load" and select the file golddust.set and click "Open". Now you can run the test.
 

 
Reshetov:

Almost every month there's a thread on this forum where another whiner, gundos, or flouderast claims that technical analysis is self-defeating.


To disprove this hypothesis once and for all, I developed and adjusted a program that performs optimization on two sections of historical data and runs a lousy test on the third section where no optimization was performed. The probability of a successful test outside the optimization sample is over 90%. At least, no matter how many times I've tried to race EURUSD, I've never waited for a single plum test.


.......



What does TA have to do with it if you are talking about optimisation? TA, as far as I know, means the application of detected patterns/models in different variations, i.e. what the process itself generates. What does this have to do with optimisation? Optimisation is about fitting. Tell me, how many people here make money using optimization?

 
NTH:


What does TA have to do with optimisation? TA, as far as I know, refers to the application of detected patterns/models in different variations, i.e. what the process itself generates. What does this have to do with optimisation? Optimisation is about fitting. Tell me, how many people here make money by applying optimisation?

What is the difference between "application of detected patterns/models in different variations, i.e. what the process itself generates " and "optimization"? In other words, how is optimization not "application of detected patterns/models in different variations"?
 

the topic seemed interesting...

tried running the program.... I tried it but...for some reason it automatically sets a stop loss of 500 and my attempts to change/reduce it did not lead to anything, it still sets 500 and...accordingly...loses

i have not had the same problem?


yes...another thing...at stage 3 the program writes that the test is unprofitable and starts closing and opening the terminal in a loop
Reason: