Jurik JMA

[just21]

Is anybody using this moving average? What are your experiences?

 

[metooxx]

works ...

 

[just21]

Do elaborate!

 

[DT-waw]

Is it really better than EMA, WMA? My friend did some tests on EOD data on many markets, many parameter sets. Performance of trading systems based on JMA crossover vs EMA crossover was similar, on average.

 

[Walther]

Quote from DT-waw:

Is it really better than EMA, WMA? My friend did some tests on EOD data on many markets, many parameter sets. Performance of trading systems based on JMA crossover vs EMA crossover was similar, on average.

Which was a best performing MA crossover method ? My guess is 5x15 or 10x35 .
Walter

 

[DT-waw]

I don't know, I don't have detailed statistics from his research. One thing I remember from my EOD backtesting is that the best set of param. for DAX index was almost the worst set for S&P ! Of course, within some range of values e.g. fast EMA 8 to 40, slow EMA 45 to 400. I don't believe in wonderful indicators or some magic set of parameters which are "the best". However, certain range of parameters can work significantly better than the others, due to market's characteristics.
I also found that swing trading gives potentially better performance than day- or long-term trading. Reasons, IMO: in daytrading we have too much noise and too high execution costs, long-term trading captures inefficiencies less frequently than swing strategies.

 

[metooxx]

Quote from just21:

Do elaborate!

We use it instead of an EMA in scalping systems, significantly better ...

 

[just21]

Has anybody got the capabailty to test different parameters of a sma crossing an ema on stock index futures of short duration?

 

[DT-waw]

Quote from metooxx:



We use it instead of an EMA in scalping systems, significantly better ...

In my view, markets at tick or 1-minute level aren't far away from random. Am I wrong? How can you profit from it? Maybe you can predict intraday volatility with some edge. I assume we talk about directional trading. I don't know what's the source of edge in scalping systems.

 

[damir00]

not even close.

Quote from Walther:



Which was a best performing MA crossover method ? My guess is 5x15 or 10x35 .
Walter

it's 2x1. i'm sure somebody can post a link to the most recent study of this...