# Elite indicators :) - page 442

151105

In the December 2011 issue of TASC there is an article about "Adjustable bands Z test". According to the author, adjustable bands are calculated as :

ABZ’ = MA (Efficacy ratio) + Stdev(Efficacy ratio)

where the explanation for "Efficacy ratio" is the following :

For the efficacy ratio, I use a ratio of the 34-day standard

deviation for the long-term indicator over a six-day standard

deviation for the short-term indicator

However I tried to use that described "Efficacy ratio" it was not giving any meaningful result in forex (maybe it works for stocks, but I was after a "universal" indicator that works on all). So, since I was failing to make that one decided to make something similar, but the other way : model for adapting part is taken from Perry Kaufman's efficiency ratio (modified for the purpose of this indicator) and bands are not built by using standard deviation but average true range.

Each part is adaptive : middle line becomes a kind of adaptive ema while average true range is adaptive in calculation too, so it is an "all adaptive" indicator. In this particular indicator I made a choice to make it "trendy" instead of fast (usually adaptive is associated with making something faster, but in this case I wanted to keep speed when "trend" changes and then keeping it smooth when the trend is steady) It is achieved with very fast calculating periods adapting (basic ema calculation allows fractional periods, and that it was perfect to use in this one). Here is an example how the calculating periods change :

and for the same periods from above, see how smooth the resulting atr channel is :

PS: it can be used as a breakout indicator, depending on the atr multiplier. I used 1 "adaptive average true range" as default channel bands since it seems to give good results for breakout way of trading

Files:
atr_ch_1.gif 23 kb
atr_ch_2.gif 24 kb
159

period converter x 3

I enclose the standard period converter script. I put it in the 'experts' folder and use as an expert advisor to produce an offline chart (e.g., 3H). Is it possible to change the code so that it will produce three offline charts instead of one?

Files:
2970
Hercs

A small hiccup..

ValeoFX

Here you go

Just one explanation : which kind of arrows you are going to get depends on alertsOnZeroCross parameter. If it is set t true, arrows on macd zero line cross will be displayed. If it is set to false then arrows on macd - signal line crosses will be displayed. Gaps control include3d already

regards

===============

Sorry Mladen, entirely MY fault! I do apologise for the false alarm. I deleted the entire request regarding the Arrows.

Thanks again.

151105

...

wojtek.paul

I am afraid that due to the way it work (an endless loop in which it writes the values) it can not be converted to do more than one offline chart (metatrader must do the "time sharing", which means you have to open 3 charts and attach the script to each with different settings in order to get 3 offline charts)

wojtek.paul:
I enclose the standard period converter script. I put it in the 'experts' folder and use as an expert advisor to produce an offline chart (e.g., 3H). Is it possible to change the code so that it will produce three offline charts instead of one?
159

28

ATR Risk calculater based on adaptiv ATR

Mladen, is it possible to make a ATR risk calculator based on adaptive ATR?

Thanx a Lot

Nod

2078

Ali Bomaye

Mladen,you for sure fly like a butterfly,but don´t sting like a bee....And you know what I mean.

I requested a mod/fusion of 2 indicators a few days ago,I sent you an e-mail with further explanations,I would have understood the following replies:No,Not Yet,Yes in a few days,I am busy,Send me more proof it may work,whatever....

No reply? I don´t understand it....Don´t worry just say no,tempting isn´t it ? Go ahead or explain if there are any cultural matters that prevent you from acknowledging a request,even to say no.

S

7

Henderson Filter

Mlande thanks for your great work, it is really valuable and helped me a lot.

I would like to to ask you a question regarding the Henderson filter.

I found the indicator herein attached somewhere, I do not remember if here or in an other site (it seams very good, may be too good to be true) The questin is : Does it repaint ?

Can you explan where to look for in the code for understanding if an indicator repaints or not ?

Is it possible to change the code to do not have the repainting without loosing too much the benefits of such indicatore?

Thanks

Gia

Files:
151105

...

Simba

Market statistics indicator need high, low, open and close prices for calculation. Tick indicator generates (and is able to generate) only one value, hence it can not be used in that indicator.

Possible solution is to use the "tick data" indicator (you have access to it) since it generates a "regular" history offline data and then apply market statistics to that offline chart. Not sure if it (market statistics) will work on an offline chart (since it depends on starting date) but it is worth to try it like that

SIMBA:
Mladen,you for sure fly like a butterfly,but don´t sting like a bee....And you know what I mean.

I requested a mod/fusion of 2 indicators a few days ago,I sent you an e-mail with further explanations,I would have understood the following replies:No,Not Yet,Yes in a few days,I am busy,Send me more proof it may work,whatever....

No reply? I don´t understand it....Don´t worry just say no,tempting isn´t it ? Go ahead or explain if there are any cultural matters that prevent you from acknowledging a request,even to say no.

S
151105

...

Gia

Henderson filter is a kind of a centered filter-average, which means that it has t extrapolate missing data from the right side. So,, yes, it recalculates half period bars (in this case I say recalculate since it has to extrapolate missing data)

Maybe the best place to read about Henderson filter is here : http://www.abs.gov.au/websitedbs/d3310114.nsf/51c9a3d36edfd0dfca256acb00118404/5fc845406def2c3dca256ce100188f8e

As of if it can be made not to recalculate : almost all centered averages are very similar to linear weighted average if the centered part is left out. If you compare the 2 you will see that the loss would be significant even though the weight coefficients would be somewhat different - the result would not be as smooth as the original filter

Gia:
Mlande thanks for your great work, it is really valuable and helped me a lot.

I would like to to ask you a question regarding the Henderson filter.

I found the indicator herein attached somewhere, I do not remember if here or in an other site (it seams very good, may be too good to be true) The questin is : Does it repaint ?

Can you explan where to look for in the code for understanding if an indicator repaints or not ?

Is it possible to change the code to do not have the repainting without loosing too much the benefits of such indicatore?

Thanks

Gia