CoeffoLine_Hist - page 3

 

Beluck,

I just posted before I got a chance to read your post. Thanks much, I will go over that one aswell. Million thanks to you.

-secXces

 

Beluck,

Going thru the code I noticed only one signifcant change. I have highlighted them in bold.

(FIXED VERSION)

ZYVar=ZYVar+(High[shift+cnt-1]+Low[shift+cnt-1])/2*(ndot+1-cnt);

TYVar=TYVar+(High[shift+cnt-1]+Low[shift+cnt-1])/2;

ZIndicatorVar=ZIndicatorVar+iMA(NULL,0,5,3,MODE_SMMA,PRICE_MEDIAN,shift+cnt-1)*(ndot+1-cnt);

TIndicatorVar=TIndicatorVar+iMA(NULL,0,5,3,MODE_SMMA,PRICE_MEDIAN,shift+cnt-1);

(ORGINAL VERSION)

ZYVar=ZYVar+(High[shift-cnt+1]+Low[shift-cnt+1])/2*(ndot+1-cnt);

TYVar=TYVar+(High[shift-cnt+1]+Low[shift-cnt+1])/2;

ZIndicatorVar=ZIndicatorVar+iMA(NULL,0,5,3,MODE_SMMA,PRICE_MEDIAN,shift-cnt+1)*(ndot+1-cnt);

TIndicatorVar=TIndicatorVar+iMA(NULL,0,5,3,MODE_SMMA,PRICE_MEDIAN,shift-cnt+1);

Beluck,

Have you seen this on a chart live? Does it do the same as the Original, like loop itsef or change its history?

-secXces

 

I just realized when looking over my last post, the bold areas did not appear. Basically it just the code below:

shift+cnt-1

shift-cnt+1

The + and - are different in the two codes. There are different on five different occasions. Two In the frist line, Two in the second line and Once again in the last line. I dont think this will have anythng to do with loop but I maybe be wrong. This seems just to shift the indicator over.

So my journey continues...lol

-secXces

 

Secxces,

I don`t think you understand what "looping" is.

It`s when the time to calculate the bars in the code takes

longer than 1000ms.

Your processor determines how fast this is done.

The fix in the mt3 code is pretty simple:

change Beginbars to 1000 (or less if you have a slow puter).

Inputs: BeginBars(1000)

 

Hey Bill,

I see what your saying. I dont know the actuall term for what the indicator is doing when it like changes its history. When i seen the fix in the mt3 version I assumed it was something to do with a loop. LOL. And then I keep using the word "retrace" which doesnt even seem to really describe what it was doing, but i think you got the idea. Can you fix the MT4 coeffofline_hist? you seem very knowledgable.

Thanks

-secXces

 

-Quick Note-

I have watched "coeffofline_hist" on a live chart. And "YES, IT DOES CHANGE ITS HISTORY ON NEW BARS, MAJOR CHANGES ASWELL". From what I have seen and the way it operates, I really dont think it is even possible to emulate the way it looks in backtesting or in its history, etc. I kinda seen this when I was looking into the indicator before, something I had seen in the history werent possible, I mean the indicator couldnt have predicted the big drops and such. So, I have come to a conclusion, this indicator officially sint a good one in my opinion. lol. 24 Hours, 13 Emails, 9 Posts, 22 Translations from Russian to English, 3 Conversations with programmers, and One Customer Suport Session later, I find myself in the same place. Good Trading to all this week.

**May the pips fall from the Heavens and land in all of your Accounts**

- secXces

 

Belluck,

Have you seen this on a chart live? Does it do the same as the Original, like loop itsef or change its history?

Tom

 

Here is something that might add some light at the "CoeffoLine" indicator


Roughly it corresponds to a linear regression slope of a 5 period smoothed (high+low)/2 (median price). I say roughly because the intention of this indicator was not to replicate the CoeffoLine indicator.

This indicator shows linear regression slope of smoothed price (price smoothing used in this one is a kind of linear weighted moving average) Here is a comparison of a 10 period regression slope (lower) and ndot 10 CoeffoLine indicator values (upper). And no, this one does not "loop into itself" nor does it repaint

 
mladen:
Here is something that might add some light at the "CoeffoLine" indicator


Roughly it corresponds to a linear regression slope of a 5 period smoothed (high+low)/2 (median price). I say roughly because the intention of this indicator was not to replicate the CoeffoLine indicator.

This indicator shows linear regression slope of smoothed price (price smoothing used in this one is a kind of linear weighted moving average) Here is a comparison of a 10 period regression slope (lower) and ndot 10 CoeffoLine indicator values (upper). And no, this one does not "loop into itself" nor does it repaint

So you mean, Linear regression slope smoothed doesn't repaint?

 
clon_tron:
So you mean, Linear regression slope smoothed doesn't repaint?

clon_tron

No. It does not repaint. You can freely use in signaling mode

Reason: