Any rookie question, so as not to clutter up the forum. Professionals, don't pass by. Nowhere without you - 6. - page 56

 
yan720:

I don't know why, but after another re-installation, the timeseries access was restored. For the period 01 to 05.07.13 it worked (it didn't work on Sunday). However, the access to the current week still didn't work correctly.

This is an update of sorts, but the original question in general still stands.

Check exactly the weekly history of W1!
 
chief2000:
When working with arrays, we often have to delete some elements of an array and shift the rest to fill those gaps.
I was wondering if there is a ready-made function that simplifies the whole process?

https://forum.mql4.com/ru/11287/page56#114804 ArrayDeleteInt() function from KimIV. Here is the whole list https://forum.mql4.com/ru/38949.

Please tell me, using a simple example, the algorithm for calculating the EMA. When I try to calculate it, like

EMA = Pri Today * K + EMA yesterday * (1 - K),

where Pris Today is today's closing price;
Yesterday's EMA;
K = 2/(N+1) values are different from terminal ones.

 
fenik:

Please tell me, using a simple example, the algorithm for calculating the EMA. When I try to calculate how

EMA = Pri Today * K + EMA yesterday * (1 - K),

where Pris Today is today's closing price;
EMA Yesterday - yesterday's EMA;
K = 2/(N+1) values are different from terminal values.


Thank you, this looks like the right one!

Perhaps this is what you've been looking for -
https://www.mql5.com/ru/forum/123182
 
chief2000:

Thank you, this looks like the right one!

Perhaps this is what you've been looking for -
https://www.mql5.com/ru/forum/123182

Thank you, it works.
 

I'm a newbie, I'm a newbie... :)

If we are talking about indicators based on neural networks, what is the best way to filter a chart before displaying it to the indicator?

I don't want to put pure data with all noise, it's ok. Differences are better than absolute values, too.

But what about noise suppression and smoothing - are there specific algorithms that are the best for this task?

Or maybe the available "standard" indicators are also good? They also take into account price changes and give fairly smooth charts, while retaining the trends of the original chart.

 

I'm really screwed here, I just can't figure it out

i have 100 orders with different candlesticks, one candlestick may contain from one to 15 orders

I want to show profit per trade ... no way...

 
VOLDEMAR:

I'm really screwed here, I just can't figure it out

i have 100 orders with different candlesticks, one candlestick may contain from one to 15 orders

I want to print profit details for each candlestick... I still have no info...

What is the problem?

 
webman1988:


Fibonacci levels are based on the percentage between points A and B, hence the formula: A-B * by the required percentage according to the Fibonacci (if the required level is 38.2 then multiply by 0.382 if the level is 61.8 then multiply by 0.618, etc.) + B

In fact, ask Google, it knows a lot if not everything)))


Thank you very much. I'm starting to understand what to do already.
 
VOLDEMAR:

I'm really screwed here, I just can't figure it out

i have 100 orders with different candlesticks, one candlestick may contain from one to 15 orders

I want to print profit statistics for each Medzhik ... I still have no info...

I want to view the profit per trade after all the other filters required:

If Magic==mn1 --> ProfitMN1+=(OrderProfit()+OrderSwap()+OrderCommission());

If Magic==mn2 --> ProfitMN2+=(OrderProfit()+OrderSwap()+OrderCommission());

etc. ...

If MagicN==mnN --> ProfitMN_N+=(OrderProfit()+OrderSwap()+OrderCommission());

How about this?

 

What about arrays?

We read the list of orders into a two-dimensional array, the first dimension is a magik, the second is a profit.

ArraySort() - the orders are lined up by magicians in order, we start summing them up and output.

Reason: