# T3 - page 32

8

If you are talking about T3 stochastic, it is a "classical" stochastic smoothed using T3 (stochastic as well as signal values are smoothed using T3 - either the original T3 (Tim Tillson) way or faster (Fulks/Matulich) way) But if you are talking about Stochastic of T3, it is a different story and in that case Stochastic is calculated using T3 filtered prices instead of "raw" prices

Thank you for the explanation and helping to clear things up.

216614

Absolute strength that can calculate values using adaptive T3 posted at this post : https://www.mql5.com/en/forum/174385/page89

Files:
8

I am a non MT4 user and am trying to code the Fulks/Matulich version of the T3 for my platform (Investor/RT) which already has a standard Tillson T3 moving average.

After explaining the difference to a buddy who was helping me with the code, he said why not simply use a Tillson's T3 with the period length cut in half in order to mimic the Fulks/Matulich version. So according to his advice, if I wanted to use a Fulks/Matulich with a length of 20, I would instead use Tillson's with the length set to 10 (or really 10.5).

The above rationale seems to be off since my guess is a Fulks/Matulich set to 20 and a Tillson set to 10.5 would not be identical.

Could someone explain what it is that I am missing with all of this? I have reviewed the MT4 code numerous times, but since my programming skills are average at best and I don't use MT4, I obviously am not understanding something.

My apologies if this is an overly amateurish questions or if it has been covered elsewhere on the forum. All help and input is veru much appreciated!

216614

raisingthebar411:
I am a non MT4 user and am trying to code the Fulks/Matulich version of the T3 for my platform (Investor/RT) which already has a standard Tillson T3 moving average.

After explaining the difference to a buddy who was helping me with the code, he said why not simply use a Tillson's T3 with the period length cut in half in order to mimic the Fulks/Matulich version. So according to his advice, if I wanted to use a Fulks/Matulich with a length of 20, I would instead use Tillson's with the length set to 10 (or really 10.5).

The above rationale seems to be off since my guess is a Fulks/Matulich set to 20 and a Tillson set to 10.5 would not be identical.

Could someone explain what it is that I am missing with all of this? I have reviewed the MT4 code numerous times, but since my programming skills are average at best and I don't use MT4, I obviously am not understanding something.

My apologies if this is an overly amateurish questions or if it has been covered elsewhere on the forum. All help and input is veru much appreciated!

The exact ratio would be the following : Fulks/Matulich length = 2*Tillson length -1 or Tillson length = (Fulks/Matulich length+1)/2

They are going to be identical if the t3 calculation you have allows fractional periods for calculation (which is often not a case)

216614

raisingthebar411:
I am a non MT4 user and am trying to code the Fulks/Matulich version of the T3 for my platform (Investor/RT) which already has a standard Tillson T3 moving average.

After explaining the difference to a buddy who was helping me with the code, he said why not simply use a Tillson's T3 with the period length cut in half in order to mimic the Fulks/Matulich version. So according to his advice, if I wanted to use a Fulks/Matulich with a length of 20, I would instead use Tillson's with the length set to 10 (or really 10.5).

The above rationale seems to be off since my guess is a Fulks/Matulich set to 20 and a Tillson set to 10.5 would not be identical.

Could someone explain what it is that I am missing with all of this? I have reviewed the MT4 code numerous times, but since my programming skills are average at best and I don't use MT4, I obviously am not understanding something.

My apologies if this is an overly amateurish questions or if it has been covered elsewhere on the forum. All help and input is veru much appreciated!

PS: you can use the one from this post https://www.mql5.com/en/forum/173058/page19 (set the adapting period to 1) and then you can enter fractional periods and you can check that they will be exactly the same when the above ratios for calculating lengths is applied

57

The exact ratio would be the following : Fulks/Matulich length = 2*Tillson length -1 or Tillson length = (Fulks/Matulich length+1)/2 They are going to be identical if the t3 calculation you have allows fractional periods for calculation (which is often not a case)

This case reminds me much about relations between Ema and Smma. Could you please inform me about other cases among modes of MA ? I'm searching via internet but it's hard to find information about this style of equivalent.

fareastol

148

Would it be possible to make a Histogram version of this indicator?

Thank you.

216614

michaelB:
Would it be possible to make a Histogram version of this indicator?

Thank you.

michaelB

Here you go

Files:
8

The exact ratio would be the following : Fulks/Matulich length = 2*Tillson length -1 or Tillson length = (Fulks/Matulich length+1)/2 They are going to be identical if the t3 calculation you have allows fractional periods for calculation (which is often not a case)

Thanks for the response.

So in order for there to be a final difference between the two, the caluclation use must not use/allow fractional periods? Is this correct?

The calculation I was working off of (more appropriately TRYING to work off of) was one along these lines:

T3(n) = GD(GD(GD(n)))

GD(n,v) = EMA(n)*(1+v) - EMA(EMA(n))*v

With n=period and v=volume factor (amp/hot).

a big thanks again for the insight!

RTB

216614

raisingthebar411:

Thanks for the response.

So in order for there to be a final difference between the two, the caluclation use must not use/allow fractional periods? Is this correct?

The calculation I was working off of (more appropriately TRYING to work off of) was one along these lines:

T3(n) = GD(GD(GD(n)))

GD(n,v) = EMA(n)*(1+v) - EMA(EMA(n))*v

With n=period and v=volume factor (amp/hot).

a big thanks again for the insight!

RTB

raisingthebar411

No

The opposite : it must be able to calculate fractional period T3 and then periods calculated using those ratios are going to give exactly the same values for T3.

That is why I told that you can use the adaptive T3 from this post https://www.mql5.com/en/forum/173058/page19 with adaptive period set to 1. Because with adaptive period set to 1 there will be no adapting and that indicator accepts fractional values for T3 calculation length