Why are quants afraid of Mark Jurik?

[bluelou]

Quote from kut2k2:

So you have nothing constructive to say about the test results. Yeah, I didn't think so. You must be psychic also, to think you know my P&L statement. We are talking about indicators in this thread, not trading strategies or bank accounts. If you want to talk strategy, go to the appropriate forum. You insinuated my AMA was crap, now you want to walk away sniping after being shown wrong. You must take lessons from imbecilebill.

BTW if you're capable, explain how to use wavelets for real-time filtering. I've seen the claims also, but they are all remarkably void on details.

Don't get all sensitive on me. I didn't say anything bad about your AMA. It might be the best AMA on ET. But, c'mon a small in-sample backtest isn't indicative of anything. I'm doing my best to take you seriously.

No, I didn't say I use wavelets for real-time filtering. I said I've tried them. I couldn't find a strategy where denoising worked for me.. That doesn't mean it doesn't work for someone else.

 

[kut2k2]

Quote from bluelou:

Don't get all sensitive on me. I didn't say anything bad about your AMA. It might be the best AMA on ET. But, c'mon a small in-sample backtest isn't indicative of anything. I'm doing my best to take you seriously.

This is ridiculous. You asked for test results and I gave you test results, using a test that was set up by an impartial third party (goodgoing) and specifically challenged to me by someone who is clearly antagonistic to me (imbecilebill). So you can hardly claim bias on my part as to the sample selection. And my indicator flew with flying colors, but suddenly it's "not indicative of anything." Riiiiiiiight.

Quote from bluelou:

No, I didn't say I use wavelets for real-time filtering. I said I've tried them. I couldn't find a strategy where denoising worked for me.. That doesn't mean it doesn't work for someone else.

You claimed real-time wavelets have been used for at least the past 8 years. If you haven't used them for real-time filtering, why bring them up at all? Denoising the past is a waste of time for a trader. No wonder you gave up on wavelet denoising. The question remains, why didn't you try the alleged real-time version? That would have been the appropriate thing to do. Do you not understand it? Did you find out it wasn't what it was touted to be?

 

[bluelou]

Congrats. I'm glad you passed your test.

 

[kut2k2]

Quote from bluelou:

Congrats. I'm glad you passed your test.

Thanks but it wasn't my test, it was somebody else's test. I already knew my AMA was better than any publicly disclosed MA. That's partly what I was trying to get across in this thread: limiting your TA to the public domain stuff is a mistake; there's better stuff out there, but you have to dig for it (or pay Mark Jurik for it ).

 

[olsen-yersen]

Quote from olsen-yersen:

The Jurik smoothing includes 3 stages:


1st stage - preliminary smoothing by adaptive EMA:
MA1 = (1-alpha)*Price + alpha*MA1[1];
2nd stage - one more preliminary smoothing by Kalman filter:
Det0 = (Price - MA1)*(1-beta) + beta*Det0[1];
MA2 = MA1 + PR*Det0;
3rd stage - final smoothing by unique Jurik adaptive filter:
Det1 = (MA2 - JMA[1]) * (1-alpha)^2 + alpha^2 * Det1[1];
JMA = JMA[1] + Det1;
where:
- Price - Price Series
- alpha - dynamic factor(will be described below)
- beta - periodic ratio = 0.45*(Length-1)/(0.45*(Length-1)+2)
- PR - Phase Ratio: PR = Phase/100 + 1.5 (if Phase < -100 then PR=0.5, if Phase > 100 then PR=2.5).



The Dynamic Factor is periodic factor (beta) raised to a power

(pow):
alpha = beta ^ Pow,
where:
- pow = rVolty ^ pow1
- rVolty - relative price volatility
- pow1 - power of relative volatility with following formula:
pow1 = len1 - 2 (if pow1 < 0.5 then pow1 = 0.5),
where len1 - additional periodic factor:
len1 = Log(SquareRoot(len))/Log(2.0) + 2 (if len1 < 0 then len1 = 0).
Thus you can see that the Dynamic factor is based on the relative price volatility giving the required
adaptability for this kind of the price filter.
The formula for relative price volatility is
rVolty = Volty/AvgVolty
(if rVolty > len1^(1/pow1) then rVolty = len1^(1/pow1), if rVolty < 1 then rVolty = 1),

JMA has already been revealed by an Ukranian developer and it has even been improved by other coders:

and this is the jurik volty stop:

 

[bluelou]

O.Y.,
What do the different colored lines represent? Since you've gone this far can you post the transformed JMA data along side the AUDUSD price info?

 

[trackstar]

please post revised Jurik code.

 

[kut2k2]

Quote from olsen-yersen:

JMA has already been revealed by an Ukranian developer and it has even been improved by other coders

OK, put it to the backtest challenge that was issued to me earlier in this thread. Your results will be interesting to see.

 

[bluelou]

Quote from kut2k2:

OK, put it to the backtest challenge that was issued to me earlier in this thread. Your results will be interesting to see.

No, don't run a backtest. That doesn't make any sense. You want the RMSE of the various filters against a benchmark instrument.

 

[kut2k2]

Quote from bluelou:

No, don't run a backtest. That doesn't make any sense. You want the RMSE of the various filters against a benchmark instrument.

The backtest makes perfect sense. It shows how well the proposed filter does against a benchmark filter (the simple moving average). How would a RMSE be better?