Playing with the VMAR`s open research

[walterw]

Here I open this new paralel thread so we can play arround freely with some cool nice ideas that vmar`s actually could suggest...

I start this thread inspired on Pyenner`s work coding the Vmar on Mt4 on a way some thought we could not get there... it had been surprised that even this last version of Pyenner work has outperformed the original VT version...

As he already presented some charts that got me thinking and some posts about possible variations, I know he can play arround freely here as myself and other traders can suggest new inovative aplications of this variations...

In my case I already had been doing some research related to scalping with this indicators, then I droped that research as did not have yet Pyenner`s last version of vmar... NOW I am more motivated to pick up again this paralel research and play here on this thread...

SO... let the creativity and interaction start... enjoy ¡¡ lets have fun ¡¡

hope you guys when trading succesfully remember to give back what God gives us with such generosity... cheers Walter.


Another day on paradise (Phill Collins)...

+ YouTube Video

[PYenner]

Some ramblings as I try to make sense of a new toy.

Walter found a VMA indicator made by Bemac for the VT platform and had the sense and experience to see that it offered something special.

Walter likes to keep trading simple (otherwise you go nuts!) and the flat lines from Bemac's indicator give a very clear reference mark on a chart for beginning and ending trades and for thinking about your options.

But what makes the indicator line go flat?
What makes it switch from following a trend to going flat and then back to trending again?

What needs to be improved in the way it-
1. follows trends?
2. changes from trend to flat?
3. changes from flat to trending?
4. does it have more uses still?

I think of it as an ADX_VMA because it comes in two parts.

The first part of the code is an ADX and some of the advantages come from the way an ADX works. It automatically adjusts itself to any range of sawtooth and ignores it. Instead, it aims at seeking out any trend component hidden in all sorts of choppy rubbish. This is what makes the flat area or horizonal effect possible. When the price goes into a choppy sawtooth the ADX main signal goes low, 0-40% is low. The VMA indicator switches from following the trend to flat line.

When a small trend is present the flat region may step up or down a few pips to show that a small trend is happening inside the sawtooth.

Once the trend component is big enough compared to the sawtooth component, the ADX gives a bigger signal, 40% up to 70% on a clean trend with little fallback. This lets the indicator switch to trend following.

There are adjustments that can be made to how closely it follows a trend and to how cleanly it switches from trend to flat and back again.
What you don't get and maybe don't need is an adjustment for the range of sawtooth movement it should ignore, that part happens automatically inside the ADX and it makes things easy.

Nice stuff and it is the ADX that does it. But if you ever tried to trade off an ADX signal you probably went nuts and gave up, too tough.

Thats where the VMA part comes in and gives a flat line when the ADX signal is below 40%.
VMA means Veriable Mean Average but what it does is simple.
When the ADX signal is low, it uses a long slow average, many bars or "periods".
When the ADX signal is high, it uses a short fast average, as few as two bars effectively.

Think of the fantail, the magenta line is the slowest to curve, it has the longest slowest curve, maybe 100 bars in that average, old news.
The Yellow side of the fantail is fastest, bends quickest, follows the trend closest, maybe 2 or 4 bars in that average, recent news only.
The VMA part decides how many bars to include in its calculation of an average price, many bars give an older price, fewer bars give a recent price.
It can stick with the past or stick with the present or somewhere in between, variable.

Trend following uses the prices from just the last few bars and ignores the past.
Flat line uses the past prices and ignores the present until it has something to say thats worth hearing.

---------------

Walter has experience at trading with the indicator and historical charts only tell you part of the story about what you need from an indicator in order to trade easily with it. So I will rely on his comments and introduce concepts for tuning ADX_VMA performance by using fewer or more "periods" or "bars" or "counts" in a sample or average. Fast means small numbers, slow means big numbers.

If the ADX samples too few bars, maybe 4 or less, you get overshoot, it follows the trend after it should have gone flat.
If the ADX uses too many bars, you get undershoot, a slow curve into or out of the flat zone, it should have gone flat quicker than that.
Optimum is somewhere in between.
Time to find a picture.

[PYenner]

Recycled one of Walters pix.
Undershoot, overshoot, what they look like, how to tune for optimum shoulder.

[walterw]

Excellent explanation there Pyenner ¡¡ ... thats a great description of the intrinsic reason of why we got an indicator that its unusual and with tremendous edge... certainly the great dilema of TA is the fact that most analist are not able to dicern (at least on the right time) the shift between cyclical and trending conditions... its the big dilema of most traders...

Now having this tool in hand you can clearly see the shift from one stage to another...

What has powerfully called my attention on this last version Pyenner (and I would suggest numbering the versions from here on ) is the responsivnes of this black line when it starts to move in a direction after being horizontal for some time... it somehow leads (not always) quite a bit on the move...



thats what calls my attention on what implications this could have as a timing tool even on the 5 min chart... the potential its yet not known... we will have to research thoroughly to be able to answer this... but common sense tells me that this type of clean performance its without doubt a good context to be working on... cheers Walter.

[PYenner]

Trend following.

This is a slightly different topic from the others.
There are some options to control how closely the VMA base line sticks to the trending prices.

Close in with the price gives maximum detail in the base line, is that detail wanted or a nuisance?

A slight gap between the VMA line and the price line occurs when you slow down the VMA by using more bars in the minimum sample size. This gives a smoother line when trending, easier on the eye but also means the Base line lags behind the price more, Good news or bad news?

Which trend following performance works best for trading?
Yellow (fastest), Blue (bit slower) or magenta (too slow)?

[walterw]

Let me see if I get this straight... here we have three vma "base lines" with diferent inputs plotted on this 1 min chart... could you tell me what inputs each one have ?


It is clear that the three lines are having diferent rolls and they all get combined into one particular trend trade...

Here I attach my first interpretation of the possible rolls they could have...




the magenta and cyan could establish a clean trend... while the yellow could give timing as it raises from horizontal effect (laddering)... some ideas... thinking in loud voice here... I like this interaction Pyenner... keep it coming... I also have some waco ideas... cheers Walter.

[PYenner]

	Quote:	walterw
	What has powerfully called my attention on this last version Pyenner (and I would suggest numbering the versions from here on ) is the responsiveness of this black line when it starts to move in a direction after being horizontal for some time... it somehow leads (not always) quite a bit on the move...

Numbering- yes, its chaos city here, you probably noticed

Responsiveness- yes, I'm surprised you noticed that.
Good eyes Walter.

You see Igorad kinda cheated a bit on one stage maybe.
But what he got may have been better that what Bemac was using.
So maybe Igorad went with what worked best instead of copying Bemac 100%.
Or maybe he ran into the same problem I had, MT4's ema function crashes indicators, I wasted 2 days on that #@$* incompetence arggggh.

Anyway, yes we have some options here, ways to optimize the shoulder at the exit from the flat bit into the trending bit and how closely it sticks to a trending price.
It needs more time to explore it....
Glad you see the importance of it.

[PYenner]

	Quote:	walterw
	Let me see if I get this straight... here we have three vma "base lines" with diferent inputs plotted on this 1 min chart... could you tell me what inputs each one have ? It is clear that the three lines are having diferent rolls and they all get combined into one particular trend trade...

Ok you got it right but you are also a step ahead.
Will do the first step here then come back to the second step.
They do overlap yes...

The first question is only about how close you want the black line on your VMA indicator to be to a trending price.

I tried to show you how it can be so close it gets really mixed in with the price and so far away that it is real smooth.

The close one shows detail, but is that what you want to see?
Or is it better to have a smoother line that shows trend, but ignores the detail?

I don't know which is best yet.
I am saying you can chose to have your black base line following either a detailed path like yellow, or a smoother "trendy" path like either blue or magenta.

It is something that can be chosen, so think about which ONE of the THREE lines would be best as your black line when there is a trend.

It wasn't the best picture. I have seen sometimes when comparing VT to Igorad that sometimes one had a smoother line up between ladder steps, while the other had more wiggles in it, like mini steps on the way up.
So do you want mini steps or smooth?
I don't have an opinion yet.... would like to hear others.

---------------------------

OK, now the second part IS about having multiple VMA lines on a 1 min chart.
But that needs to be discussed on its own and it has more to do with the flat bit and the shoulders going into or out of the flat bit.
I did post a picture on the other thread that introduced that line of thinking.
I want to take it a lot further because it has possibilities.

One possiblity is that, in addition to your black line which is flat, you can also have a few more lines, maybe one that works a bit faster so it doesn't stay so flat but it may give you an early warning before your black base line turns up or down into a new trend. That may not be keeping it simple, but it might still be handy.

---------------

A third possibilty is replacing the fantail we have now with a fantail of VMA lines. Now there is one scenario where this might give you an exit price guide when trading a recovery. Its too many things to think about, they need to be split up and developed. I suspect a VMA fantail may be more useful than the fantail we have now. Scalping the swings either side of the flat line may also come into it. Dude, it gets all mixed up, overload, meltdown, kablooowweee
One at a time, keep it simple, argh, I should be so lucky
Pass me a banana someone please...

[islandhome]

I use vma this morning to detect bounceof resistance namely the channel (ala SRDC 2 method) Flattening out of vma on 15 min chart was my trigger to initiate possible entry.http://www.megaupload.com/?d=7AYF9QGS. Dropped down to 5 min to observe price action. Waited for 21 sma our trendsetter ( white ma) to cross vma and then enter on the close of penetrating candle. Exited trade when 21 sma crossed over vma horizontal. .http://www.megaupload.com/?d=KTMMNK92. Thx again walter and pyenner.

[PYenner]

One possible advantage of using VMA lines for a fantail.

The slowest line sometimes (recovery) gives an estimate of what your exit price might be.

In this case magenta is the slowest line and since it gives the longest flat line during a recovery, then you have a nice fixed estimate of where your trade might end up going.

Recycled another pix.
Try to see the general idea, when the fantail diverges, all the lines will eventually converge again. In the case of a recovery, the faster lines "chase after" the slowest line, so that is where they will likely meet.
Magenta is sometimes a guide to exit price.