Download MetaTrader 5

One more average ...

To add comments, please log in or register
Mladen Rakic
61972
Mladen Rakic  

The usual story goes something like this : which is the best moving average? Everyone that ever started to do any kind of technical analysis was pulled into this "game". Comparing, testing, looking for new ones, testing ...

______________________________

The idea of this one is simple : it should not be itself, but it should be a kind of a chameleon - it should "imitate" as much other moving averages as it can. So the need for zillion different moving averages would diminish. And it should have some extra, of course :

The extras :

  • it has to be smooth
  • it has to be able to "change speed" without length change
  • it has to be able to adapt or not (since it has to "imitate" the non-adaptive as well as the adaptive ones)

______________________________

The steps :
smoothing - compared are the simple moving average (that is the basis and the first step of this indicator - a smoothed simple moving average with as little lag added as it is possible and as close to the original as it is possible) Speed 1 and non-adaptive are the reference for this basic setup.
"speed changing" - same chart only added one more average with "speeds" 2 and 3 (for comparison purposes only here)
finally - adapting : same chart with SMA compared to one more average with speed 1 but adaptive (so this parameters would make it a "smoothed adaptive simple average") Adapting part is a modified Kaufman adapting way and this part (the adapting part) may be a subject for changes in the future (it is giving satisfactory results, but if or when I find a better way, it will be implemented here)

______________________________

Some comparisons for different speed settings (all the comparisons are without adaptive turned on, and are approximate : approximation comes from a fact that it is impossible to get exactly the same values from only one way of calculation, and frankly, I even did not try to get those same values)
speed 0.5 - T3 (0.618 Tilson)

speed 2.5 - T3 (0.618 Fulks/Matulich)

speed 1 - SMA, harmonic mean

speed 2 - LWMA

speed 7 - very similar to Hull and TEMA

speed 8 - very similar to LSMA and Linear regression value

Due to some properties of this average, it is very convenient to use it as a part of other indicators for pre- (as a filter before calculation) and post- smoothing. Some of the indicators that are going to be posted in this section are going to use this way to achieve those goals

______________________________

And, the last (it might be posted before the rest, but )

Parameters :
Length- length (period) for averaging AppliedPrice- price to use for averaging (it uses the usual metatrader values :
0 - close

1 - open

2 - high

3 - low

4 - median (high+low)/2

5 - typical (high+low+close)/3

6 - weighted (high+low+close+close)/4

Speed- desired speed (i limited to -1.5 on the lower side but it even does not need that limit - some interesting results with speeds that are less than 0 can be achieved)

Adaptive- does it adapt or not

PS: did not think of a different name for it.We could name it whatever we want (since it does not exist anywhere else but here on TSD) but I rather like this name ______________________________

Updated version posted here : https://www.forex-tsd.com/forum/exclusive/11428-one-more-average/page15#comment_763153

SIMBA
2078
SIMBA  

OMA...Bands?

mladen:
The usual story goes something like this : which is the best moving average? Everyone that ever started to do any kind of technical analysis was pulled into this "game". Comparing, testing, looking for new ones, testing ...

______________________________

The idea of this one is simple : it should not be itself, but it should be a kind of a chameleon - it should "imitate" as much other moving averages as it can. So the need for zillion different moving averages would diminish. And it should have some extra, of course :

The extras :

  • it has to be smooth
  • it has to be able to "change speed" without length change
  • it has to be able to adapt or not (since it has to "imitate" the non-adaptive as well as the adaptive ones)

______________________________

The steps :
smoothing - compared are the simple moving average (that is the basis and the first step of this indicator - a smoothed simple moving average with as little lag added as it is possible and as close to the original as it is possible) Speed 1 and non-adaptive are the reference for this basic setup.
"speed changing" - same chart only added one more average with "speeds" 2 and 3 (for comparison purposes only here)
finally - adapting : same chart with SMA compared to one more average with speed 1 but adaptive (so this parameters would make it a "smoothed adaptive simple average") Adapting part is a modified Kaufman adapting way and this part (the adapting part) may be a subject for changes in the future (it is giving satisfactory results, but if or when I find a better way, it will be implemented here)

______________________________

Some comparisons for different speed settings (all the comparisons are without adaptive turned on, and are approximate : approximation comes from a fact that it is impossible to get exactly the same values from only one way of calculation, and frankly, I even did not try to get those same values)
speed 0.5 - T3 (0.618 Tilson)

speed 2.5 - T3 (0.618 Fulks/Matulich)

speed 1 - SMA, harmonic mean

speed 2 - LWMA

speed 7 - very similar to Hull and TEMA

speed 8 - very similar to LSMA and Linear regression value

Due to some properties of this average, it is very convenient to use it as a part of other indicators for pre- (as a filter before calculation) and post- smoothing. Some of the indicators that are going to be posted in this section are going to use this way to achieve those goals

______________________________

And, the last (it might be posted before the rest, but )

Parameters :
Length- length (period) for averaging AppliedPrice- price to use for averaging (it uses the usual metatrader values :
0 - close

1 - open

2 - high

3 - low

4 - median (high+low)/2

5 - typical (high+low+close)/3

6 - weighted (high+low+close+close)/4

Speed- desired speed (i limited to -1.5 on the lower side but it even does not need that limit - some interesting results with speeds that are less than 0 can be achieved)

Adaptive- does it adapt or not

PS: did not think of a different name for it.We could name it whatever we want (since it does not exist anywhere else but here on TSD) but I rather like this name

Mladen,

Once again,Thanks for your marvellous indicators.

I have been testing it with the visual tester :

EURUSD H4

SETTINGS:13,0,-1,TRUE

The OMA is very robust...And I believe that if you add bands to it(preferably ATR bands),the end result could be even better than your exceptional TMA Bands...Tma bands slightly "recalculate" ...This OMA,unless I am losing my sight,doesn`t redraw at all,with settings I posted...and,yes not all touches at the external band would be profitable entries(most will be),but we can use it to add to existing positions in a trend,or we can use them as I explained in the TMA bands thread...I have the feeling this beast is much better than TMAs,so,if you can do the bands,I will open a thread in this section and explain the several uses I would give them....Hopefully you have time to do so .

I will do a few more visual tests with different settings,but I was so enthralled by what I saw with those settings,that I repeated the test several times...Now could you explain in layman`words what is the difference between adaptive true/false?...Most of the indicators I have tested,when put on adaptive mode,look exceptional after the fact,when put in Live trading,they are not so exceptional ...So,I was surprised by the robustness OMA showed at adaptive mode....This is a jewell,believe me...But you already knew that

Regards

S

Mladen Rakic
61972
Mladen Rakic  

Simba,

You hit it in the head from the beginning (as usual ). The -1 for speed is really an interesting one (frankly, those "speeds" less than zero were what made me aware of some of the properties of this average while I was working on it)

________________________________

Of repainting : no, no repainting at all

Of adaptive : nothing new in that part applied from me (as I said, that is basically Kaufmans way of adapting, slightly modified to avoid the "bumps" of AMA) What it does is that it shortens the average period in volatile periods. So, in fact, in those times, it makes it a "shorter" (or "faster") in order to make it more responsive to changes. In the quiet market it reverses to its slowest period (the nominal period of the average)

Of developments : yes, will make some "channel" versions. Also, if you look at the cfb thread, you will see some different ideas that I will probably use in the adaptive part of this one. I somehow feel that this one deserves attention and that it can be useful in a various ways (as a smoother in indicators there are very few ones that can compare to this one, also as a filter, and since you already noticed its usefulness as a "normal' average,... )

SIMBA
2078
SIMBA  

Trend filter

mladen:
Simba,

You hit it in the head from the beginning (as usual ). The -1 for speed is really an interesting one (frankly, those "speeds" less than zero were what made me aware of some of the properties of this average while I was working on it)

________________________________

Of repainting : no, no repainting at all

Of adaptive : nothing new in that part applied from me (as I said, that is basically Kaufmans way of adapting, slightly modified to avoid the "bumps" of AMA) What it does is that it shortens the average period in volatile periods. So, in fact, in those times, it makes it a "shorter" (or "faster") in order to make it more responsive to changes. In the quiet market it reverses to its slowest period (the nominal period of the average)

Of developments : yes, will make some "channel" versions. Also, if you look at the cfb thread, you will see some different ideas that I will probably use in the adaptive part of this one. I somehow feel that this one deserves attention and that it can be useful in a various ways (as a smoother in indicators there are very few ones that can compare to this one, also as a filter, and since you already noticed its usefulness as a "normal' average,... )

Thanks for your reply and for your kind comments

Yes,this one deserves attention....Look at the attacchment...2 OMAs with settings :

RED:13,0,-1,TRUE

AQUA :13,0,-1,FALSE

Very good trendfilter for H4...we can then just trade retracements to the external OMA or wait for near perfect entries,when there is the first countertrend penetration of the OMAs by the prices...see Aqua rectangles.

Obviously,entries will be even better once we have the bands

I am still experimenting,and I have to say that number 1 is very similar to sma ,as you explained,but smoother...and,for the moment,I prefer the combo I attacched as the best trading setup for H4...the -1 setting is extremely robust.

Regards

S

Files:
oma.gif 70 kb
SIMBA
2078
SIMBA  

Retracements

Another example of retracements to the external band,used for entries,see pic...

..OMA Bands would be an excellent gift,and,since you are alreading developing them,I will wait for them before opening a thread ,while I post my thoughts here.

Regards

S

Files:
oma2.gif 66 kb
Mladen Rakic
61972
Mladen Rakic  

...

Here you go

ATR bands this time, but you can control the ATR completely through parameters (you can even turn bands off )

Now, without kidding :
ShowBands-> the switch to show or not the bands

AtrMultiplier-> multiplier for ATR (can be decimal)

AtrPeriod-> period applied to ATR

AtrShift-> number of bars ATR calculation is shifted (set to the "old" trick with atr from before 10 bars by default)

Will try some other versions of bands (% for example, I would like to see how does it behave. If I find it interesting, most definitively will post it here)

regards

mladen

SIMBA:
Another example of retracements to the external band,used for entries,see pic...

..OMA Bands would be an excellent gift,and,since you are alreading developing them,I will wait for them before opening a thread ,while I post my thoughts here.

Regards

S

updated version posted here: https://www.mql5.com/en/forum/179662

SIMBA
2078
SIMBA  

Gracias

mladen:
Here you go

ATR bands this time, but you can control the ATR completely through parameters (you can even turn bands off )

Now, without kidding :
ShowBands-> the switch to show or not the bands

AtrMultiplier-> multiplier for ATR (can be decimal)

AtrPeriod-> period applied to ATR

AtrShift-> number of bars ATR calculation is shifted (set to the "old" trick with atr from before 10 bars by default)

Will try some other versions of bands (% for example, I would like to see how does it behave. If I find it interesting, most definitively will post it here)

regards

mladen

Thankyou,merci,grazie,danke,gracias....hvala!

S

Cyclesurfer
9174
Cyclesurfer  

BB Omacd

Made this mod using OMA and BB Macd looks interesting and also OsOma!Guess could have named them Obama, but didn't want to ruin the indicator

Ps) Mladen hope you don't mind me changing this version to Oma too lazy to type out One more average

Updated OMA posted here : https://www.mql5.com/en/forum/179662Updated BB OMA macd posted here : https://www.mql5.com/en/forum/179662

Files:
bbomacd.gif 71 kb
osoma.mq4 3 kb
oma.mq4 4 kb
Mladen Rakic
61972
Mladen Rakic  

...

Good work tools, I like it And no, I do not mind the OMA (I am lazy too, so let's call it OMA)

mrtools:
Made this mod using OMA and BB Macd looks interesting and also OsOma!Guess could have named them Obama, but didn't want to ruin the indicator Ps) Mladen hope you don't mind me changing this version to Oma too lazy to type out One more average
SIMBA
2078
SIMBA  

request

I received a request from haubentaucher...

H,please find my reply in the pic,and I hope that it answers your q...If it doesn`t,please feel free to POST HERE why ,but do not ask me again by PM.

Regards

S

Files:
atthebest
152
atthebest  
mrtools:
Made this mod using OMA and BB Macd looks interesting and also OsOma!Guess could have named them Obama, but didn't want to ruin the indicator Ps) Mladen hope you don't mind me changing this version to Oma too lazy to type out One more average

Thnak to all fo you guys for your effort .

Would you please kindly help me in the rulse of trading sell or buy and TP & SL

Your help is highly appreciated.

kindly if you dont mind attach you template