Elite indicators :) - page 91

New comment
 
mrtools:

Hi,

mrtools or someone can post here all indicator and the telpmate required to show the graph like in the image above?

thank you very very much

 

...

823,

mrtools already posted one version of dynamic trend oscillator (dtosc) on the same post that picture came from (this post : https://www.mql5.com/en/forum/general or you can download it from here : https://www.mql5.com/en/forum/general too) The rest of the indicators are mrtools's and it is up to him if he will upload them here

regards

mladen

 
mladen:
823,

mrtools already posted one version of dynamic trend oscillator (dtosc) on the same post that picture came from (this post : https://www.mql5.com/en/forum/general or you can download it from here : https://www.mql5.com/en/forum/general too) The rest of the indicators are mrtools's and it is up to him if he will upload them here

regards

mladen

DTOSC I know...

...but I'm interested at the rest of the indicators.

Someone can post it?

Mrtools?

Thanks!

 

Singular Spectrum Analysis ...

mrtools knows me better that the majority, so he will laugh when I say that one line of code that goes like this :

X = Low + (High-Low)/2;

made me post these two messages

______________________________________

There was quite a lot of talk on public part of the forum about Singular Spectrum Analysis (simbas thread with a lot of information about it, some of the posts concerning this matter starts roughly from here https://www.mql5.com/en/forum/175938 (if I did not pinpoint the start exactly I apologize)) and for some more general info of what exactly SSA is you can find here Singular spectrum analysis - Wikipedia, the free encyclopedia )

______________________________________

Here you can find two "flavours" of SSA : a "simple" one

and one with atr channel added (btw: ATR channel parameters are correct : the idea is to use past ATR rather than the current one to "predict" future one, hence the "+10" in the bars tested for ATR)

Also, you can find the C/C++ source (it is compilable in both, no C++ specifics in it) source of the library that is used to calculate SSA. I use CodeBlocks for that job so someone using other one will probably need to create a project of its own) linSSA.dll should be copied to experts/libraries folder of metatrader in order that SSA work as it should

Files:
ssa_-_price.gif  17 kb
ssa_-_price_bands.gif  18 kb
libssa.zip  171 kb
ssa_of_price.mq4  3 kb
ssa_bands.mq4  4 kb
 

Singular Spectrum Analysis ... continued

What I would like is to draw attention to one obvious similarity (result vise, as I already stated in public part of forum) of SSA and centered Triangular MA.

Here are the implementations similar to SSA one : triangular MA centered (with a little addition that makes it different than the one posted on public forum)

and triangular MA bands (same rules a for SSA bands)

Also attached here just the simple dll of libSSA (for those that are not interested in C source of the SSA library)

 

Ssa

Too cool Mladen, like the new TMA, when was first enlightened about SSA tried it on everything, check out WPR first one smoothed with SSA now bottom one same WPR Period with reg. SSA, the smoothed still crosses oversold and overbought quicker. The dots on the squeeze are with SSA and the histo with HMA.The HMA by Mladen a few pages back.

Files:
ssaofwpr_v1.mq4  4 kb
hssa_squeeze.mq4  6 kb
ssa.gif  70 kb
 
mladen:
What I would like is to draw attention to one obvious similarity (result vise, as I already stated in public part of forum) of SSA and centered Triangular MA.

Here are the implementations similar to SSA one : triangular MA centered (with a little addition that makes it different than the one posted on public forum)

and triangular MA bands (same rules a for SSA bands)

mladen: great work. Seems to me that the triangular method tends to react more accurate and thus with less "overshoot".

Is that assumption correct? And...what's the "little addition" you're referring to?

Last question: Should this indicator be 'calibrated' daily for each pair?

Greetz!

Snow.

 

Snow.

It depends on the market conditions. In some cases SSA will be faster and in other triangular MA is going to be faster.

The answer to the second question : no need to 'calibrate' them.

________________________

PS: please do bear in mind that both of those indicators (and derivatives of them) are going to recalculate. (I am not going to use the "repaint" term since it is meaningless in cases like these : Hodrick-Prescott filter, SSA, centered MAs, ... a lot of very useful indicators work like that and their math models are like that, so they are simply calculating some values over and over. In cases of SSA it is a lag bars, in case of triangular MA centered it is HalfLengthbars)

That is the main reason why they should be used in manual trading only (some "signals" are going to be changed with time, these do need your mind as a decisive factor for entering trades)

________________________

Personally I like them (once I told to a friend of mine that "they give results so beautiful that there must be some degree of truth in them") They show what I like to call : "It would be logical if the price would go in this direction" The rest of the logic is in our feeble minds when it comes to usage of these indicators

regards

mladen

Snowski:
mladen: great work. Seems to me that the triangular method tends to react more accurate and thus with less "overshoot".

Is that assumption correct? And...what's the "little addition" you're referring to?

Last question: Should this indicator be 'calibrated' daily for each pair?

Greetz!

Snow.
 

? Lost

Dear friends,

... I feel like a little mouse in the room of titans here! LOL

Congratulations for your awesome work, really!

Just amazed for some of the indicators pictured out here, I was trying to make a bit of homework and find out something more about the "squeeze" series, because I found mrtools' quite interesting.

I went around quite a bit, but I found that mladen's "Advanced BBsqueeze" in post #175 of the MTF indicators thread... disappeared.

Is anybody so kind to give me a sort of "pathmap" to have some more information? Look, I'm a strange guy: sometimes I like also to "understand" what's behind things... . Mladen, you're often so kind to post also articles or "common English" descriptions of strategies and indicators...

Keep on with the superb discussion!

F

 

Tma

mladen:
Snow.

It depends on the market conditions. In some cases SSA will be faster and in other triangular MA is going to be faster.

The answer to the second question : no need to 'calibrate' them.

________________________

PS: please do bear in mind that both of those indicators (and derivatives of them) are going to recalculate. (I am not going to use the "repaint" term since it is meaningless in cases like these : Hodrick-Prescott filter, SSA, centered MAs, ... a lot of very useful indicators work like that and their math models are like that, so they are simply calculating some values over and over. In cases of SSA it is a lag bars, in case of triangular MA centered it is HalfLengthbars)

That is the main reason why they should be used in manual trading only (some "signals" are going to be changed with time, these do need your mind as a decisive factor for entering trades)

________________________

Personally I like them (once I told to a friend of mine that "they give results so beautiful that there must be some degree of truth in them") They show what I like to call : "It would be logical if the price would go in this direction" The rest of the logic is in our feeble minds when it comes to usage of these indicators

regards

mladen

Yes,these indicators make you think about what will be the most probable course of action,and they get you to think dually,which is the best way to think when trading .

I have learnt to enjoy the TMA centered,ever since you were kind enough to post it in one of my threads..Now,I have tried to take the new one you posted here and tried to do a mod which basically consists in "projecting" it into the future,I have used Fourier extrapolator to no avail(which I have used successfully for other indicators),the other option is to use the Polyfit indicators by igorad and try to bridge the 2 together,but I don`t even know how to start ...I am not a coder(that is obvious),so,I would like to ask you,if you think that it could be useful,if you could devise some way to do this mod,basically project your indicator(which is already centered) into the future,n bars,selectable by user,and,specially the TMA with bands.

My idea is to use if for trading,setting one at half span equal to half the span of the dominant cycle(using Corona charts you can know the dominant cycle ),and another one set to half THAT span...basically I am trying to replicate what Hurst did with his channels,and what Brian Millard and the guy at swingmachine are probably already doing...both crossovers and projected bands will probably be extremely useful in trading...basically,this method will combine both Time(half cycle) and price(std of price),and ,with practice,will help finetune our trading.

If you think it is a good idea,and you want to do it,I can open a thread here and show anybody interested how to use them.

Thanks for your indicators,as usual,they are awesome.

Regards

S

1...848586878889909192939495969798...1519
New comment
Reason: