- This topic has 227 replies, 14 voices, and was last updated 10 years, 3 months ago by
.
- Posts
http://www.ifta.org/public/files/journal/d_ifta_journal_09.pdf
Just look at White bar on Gadi_NormalizedVolume that extends beyond the avg.
Majority of them are signals for direction change.
I’m glad that slowly, but surely, almost everyone acknowledge Volume (or Ticks/Activity) as a Leading indicators.
About “R” analysis: been there, done that long time ago. I’ll be glad to see others confirming or using different approach.
G.
Hi @gg53. I’ve seen that blue and red candles contain a small white proportion. Can you give me an advice how to identify only red coloured bars in mql4. I’ve used the iCustom function
iCustom(NULL,0,"Gadi_NormalizedVolume_v2.2",8,10000,x,i+1)to access indicator values at which
x = 0refers to the MA line,x = 1refers to blue and some white coloured bars,x = 2refers to red and some white coloured bars, andx = 3refers to white only bars.
Hi @gg53. I’ve seen that blue and red candles contain a small white proportion. Can you give me an advice how to identify only red coloured bars in mql4. I’ve used the iCustom function
iCustom(NULL,0,"Gadi_NormalizedVolume_v2.2",8,10000,x,i+1)to access indicator values at whichx = 0refers to the MA line,x = 1refers to blue and some white coloured bars,x = 2refers to red and some white coloured bars, andx = 3refers to white only bars.
Use the one with “fix” extension to it @FF “Killing Zone”, or use the attached.
This one have solid color on bar close, but I use the other version to note when a Doji (white) becomes non-Doji. I can understand that this is confusing to some people.
G.
There’s also an indicator which helps to export also indicator values to csv files.
Thanks for pointing to hanover’s indicator. I’m having a different approach in mind, like an export function library that can be linked to any custom made indicator or script.
A good trader is a realist who wants to grab a chunk from the body of a trend, leaving top- and bottom-fishing to people on an ego trip. (Dr. Alexander Elder)
chek this primary version of carver osillator
shows direction of candle b4 it starts…
special thanks to buddy Anti
Looks great. Well in some cases it really shows changes in direction before. Thanks, but you did it!
just finish it.doing forward testing…
see here the direction of line is doen side but market going up it indicate market is sure coming down
we have to trade in direction of line
eg.if line showing doen direction we have to sell even price going up
vise versa
i thank u from bottom of my heart its u without whom me not able to complete the code im stuck but u solve the problem…
still working on tgt part soon will finish and distribute it….
Hmmm, this is a ‘little bit’ off topic, cluttering my own thread, but I don’t think it makes sense to start a new one for this …
The whole discussion about volume and tick data has placed a nagging bug in my brain … I just had to code it just to show it can be done. Too early to post the indicator, and maybe it’s good for nothing from a practical point of view.
What I did yesterday was to code a bunch of tick data based currency strength prototypes. The screenshot shows two instances of the latest version, one of them showing smoothed lines (Ehlers’ SuperSmoother), the other one raw data.
Both instances are running on tick data, producing 60 bars of 5 seconds timeframe synthesized from the ticks. This means, we’ve got a magnifying glass which zooms into 5 M1 bars, converting them into 60 values. The timeframe can be chosen freely to any integer amount of seconds also the window width.
Basic arithmetics is the same as in my already published CIX indicator, with all lines reset to zero at the left (past) side of the time interval.
For now, this is just a prototype, acting as a proof of concept that sophisticated analysis can be done on tick data without driving CPU load to the ceiling (processor load on my system does not even flicker when I load this indicator).
Any thoughts?
s.
A good trader is a realist who wants to grab a chunk from the body of a trend, leaving top- and bottom-fishing to people on an ego trip. (Dr. Alexander Elder)
Thanks and yes:
Try it on 5 Seconds chart (each bar contains 5 Sec.). Very informative.
1 second is possible, but much too ‘nervous’. 5 to 10 seconds seem to be a good choice to indicate where ultra short term momentum goes.
Have to try next week whether resetting the beginning of the time window to zero is a good idea.
s.
A good trader is a realist who wants to grab a chunk from the body of a trend, leaving top- and bottom-fishing to people on an ego trip. (Dr. Alexander Elder)
5 Sec. bars will give you good indication of trade entry when there is a sudden jump of 3-4 pips.
G.
tankhang just told me (via Skype) that he found a treasure, here: http://www.traderslog.com/defining-the-trend and sent me the MQ4 source (attached). This is very, very similiar to mine (difference in bar/timing calculation) .
I’d like to come back to this source posted. AFAIK, we do not have a version based on a single function for EA integration yet. I have two different versions of such a function now, following different algorithmic approaches to model the mathematical nature of the IIR smoothing with sufficient precision.
If we want to call a single function in our EA, using IIR filtering and not FIR, we either have to use a static or global array to store historic values or ensure for a certain redundant calculation inside our function (my preferred approach at the moment). In either case, we will have to define a level of precision we want to achieve, or a level of inaccuracy we’re willing to tolerate, due to the nature of the IIR filter.
Let’s say VZO is mostly swinging between +50 and -50 (true for most cases). So what’s the level of inaccuracy we’re going to tolerate here? I would say a precision of +/- 0.1 is sufficient. Does it make a difference if the VZO shows 24.1 or 24.2 points?
The smaller our accepted inaccuracy is, the more redundant algorithmic loops we will have to perform.
Alternative approach to the above mentioned: use an FIR filter instead of IIR, e.g. an ALMA. Has anybody tried that?
s.
A good trader is a realist who wants to grab a chunk from the body of a trend, leaving top- and bottom-fishing to people on an ego trip. (Dr. Alexander Elder)
[ quote=10922]
…. Alternative approach to the above mentioned: use an FIR filter instead of IIR, e.g. an ALMA. Has anybody tried that? s. [/quote] Never test it, but just got a small function that can be converted to mql4 i think : http://www.financial-hacker.com/trend-delusion-or-reality/
double ALMA(double &Data, int Period)
{
double m = MathFloor(0.85*(Period-1));
double s = Period/6.0;
double alma = 0.0, wSum = 0.0;
int i;
for (i = 0; i < Period; i++){
double w = MathExp(-(i-m)*(i-m)/(2*s*s));
alma += Data[Period-1-i] * w;
wSum += w;
}
return (alma / wSum);
}There is also a link to download the pdf, a brief information to alma calculation …
Thank you, @smallcat:
Never test it, but just got a small function that can be converted to mql4 i think : http://www.financial-hacker.com/trend-delusion-or-reality/
This is not what I meant – I coded the ALMA in mq4 a long time ago. What I meant is whether anybody has tried to apply an FIR filter like ALMA to the VZO algorithm instead of the EMA (IIR).
s.
A good trader is a realist who wants to grab a chunk from the body of a trend, leaving top- and bottom-fishing to people on an ego trip. (Dr. Alexander Elder)
Thank you, @smallcat:
Never test it, but just got a small function that can be converted to mql4 i think : http://www.financial-hacker.com/trend-delusion-or-reality/%5B/quote%5D This is not what I meant – I coded the ALMA in mq4 a long time ago. What I meant is whether anybody has tried to apply an FIR filter like ALMA to the VZO algorithm instead of the EMA (IIR). s.
I’ll respond and relate to that AFTER the basic EA skeleton in “Tradind made REALLY simple” will be finished.
G.
Thanks, G.:
I’ll respond and relate to that AFTER the basic EA skeleton in “Tradind made REALLY simple” will be finished.
So for the moment I’ll put together my EMA based version according to my above post, allowing roughly a tolerance of about 0.1 %. This can be achieved at reasonable CPU resource consumption, ans should be precise enough to trade.
s.
A good trader is a realist who wants to grab a chunk from the body of a trend, leaving top- and bottom-fishing to people on an ego trip. (Dr. Alexander Elder)
Ok, this is my experimental version of Volume Zone Oscillator.
It contains two different implementations of the VZO algorithm to deal with the effect of an IIR filter that a certain number of precalculated values are required to ensure precise results.
When experimenting with both versions, it’s recommended to set parameter ‘maximum bars’ to a low value like 50. So you can see the different outcome easily.
My final and suggested version will be posted over in G.’s Trading made REALLY simple thread for use in our community EA.
This post is meant to discuss the concept and algorithms. My currently suggested version for practical use is in G.’s thread.
As with most of my mqh libs, I skipped the effort of precompiling the lib and linking it via a header file – feel free to change!
s.
A good trader is a realist who wants to grab a chunk from the body of a trend, leaving top- and bottom-fishing to people on an ego trip. (Dr. Alexander Elder)
Unfortunately I’m still snowed under with work. Nevertheless, here’s something I’d like to share with you. It has been built based on my up/down-volume estimation introduced earlier and just uses the volume relationship for trading decisions (entering candle after arrow candle). In most cases it gives good signals over all TFs.
Unfortunately I’m still snowed under with work.
Me too: German tax authorities demand a good part of my attention for a couple of days …
My experience says if you don’t take them seriously they can affect you seriously … So I got to change my focus for a while!here’s something I’d like to share with you
Hmmm … those signals are looking quite nice at first sight, but when looking closer, in many cases there seems to be a lot of drawdown on the first candle after the signal …
Was it built with the UpVol / DnVol indicators (or similar) you posted lately last year?
s.
A good trader is a realist who wants to grab a chunk from the body of a trend, leaving top- and bottom-fishing to people on an ego trip. (Dr. Alexander Elder)
Hi @gg53. I’ve seen that blue and red candles contain a small white proportion. Can you give me an advice how to identify only red coloured bars in mql4. I’ve used the iCustom function
iCustom(NULL,0,"Gadi_NormalizedVolume_v2.2",8,10000,x,i+1)to access indicator values at whichx = 0refers to the MA line,x = 1refers to blue and some white coloured bars,x = 2refers to red and some white coloured bars, andx = 3refers to white only bars.
Use the one with “fix” extension to it @FF “Killing Zone”, or use the attached. This one have solid color on bar close, but I use the other version to note when a Doji (white) becomes non-Doji. I can understand that this is confusing to some people. G.
Thanks a lot G
… about 5 sec bar, i assume it is not using MT4 platform ? Or is there a trick to convert tick data to get <1 minute bars ?
My experience says if you don’t take them seriously they can affect you seriously … So I got to change my focus for a while!- You must be logged in to reply to this topic.