Sorry :eek: As I see it... Simple MA's enter too late and exit too late. Adaptive MA's enter too early (before a trend change is confirmed) and exit too early (on a retrace, whilst the trend is still in place). You need to find something in between. I found the Exponential meets this goal.
Jurik Research has published an article comparing different types moving averages in a certain extent. http://www.jurikres.com/down/ma_evolv.pdf
never ever any piece of information. i understand the business concept of jurik, but i am very reluctant to buy trading tools and not understand them. thanks anyways.
Try Kaufman's MA. I think that most people look at a MA as simply some component of a mechanical system without trying to understand the logic of the tool. Kaufman's MA is based on a volitility calculation to change the speed of the indicator. By changing the time length for any of the three components, you can drastically change the characteristics of the tool. Try experimenting with different values for the short and long periods. I use this tool a lot as it has virtually no slope during congestion and very little lag on breakouts. It doesn't really have much value as a S/R point like other MA's do. Try using it in creative ways. The best part of it is its price: free. I don't believe in paying for an indicator, especially if the vendor is very vague about its logic.
On Kaufman And Tillson - here is an extract from a summary I made of an article I once read on T3 (or Tx) "So, if you multiple run with twicing and turn down the volume factor (Amplitude Response) by an appropriate factor you can get what might be called a Generalized DEMA (GD) GD(n,v)=EMA(n)*(I+v)-EMA(EMA(n))*v Where v ranges between zero and 1. When v=0, GD is just an EMA, and when v=1, GD is DEMA. In between ,GD is a less aggressive version of DEMA. By using a value for v of <1 we cure the multiple DEMA overshoot problem at the cost of accepting some additional phase delay.. Now we can run GD through itself multiple times to define a new, smoother, moving average (T3) that does not overshoot the data. 6 T3(n)=GD(GD(GD(n))) Results are v similar to DEMA but smoother In filter terminology T3 is a six-pole non linear Kalman filter. Kalman Filters use the error - in this case (the time series - EMA(n)) â to correct themselves. In TA these filters are known as adaptive MAâs; they track the time series more aggressively when it is making large moves." The bit that interests me most is the "Where v ranges between zero and 1. When v=0, GD is just an EMA, and when v=1, GD is DEMA. In between ,GD is a less aggressive version of DEMA. By using a value for v of <1 we cure the multiple DEMA overshoot problem at the cost of accepting some additional phase delay" On the other hand if you traded on crossovers plotted in a separate window it seems to me that v of > 1 may well be an advantage in identifying the crossover point more clearly at no cost in lag?
HiLo rongvald You are absoluterly correct, T3 is a Kalman Filter, but it is not an ADAPTIVE AVERAGE as mind tries to claim. Neither is Mark Juricks. In 1993 M.J. was publishing a cook book on how he created his average. After that he pulled all his explanations from the market. T3 is unsuccesfull duplicate. Unfortunately what mind is describing is not adaptive average. It uses a multiplier to adjust some preset and precalculated values. Proper self-optimizing (Walking) average will not overshoot, because it does not have any mechanical coefficient that adjusts some other arbitrary average. I have posted a screen shot of self-optimizing (Walking) average
Hi, I find that KAMA is so reactive that there are virtually no periods where it is flat. I thought about what you said to see if I could improve my system entries and exits but when I look through a chart I can't see any major periods where it is trading in flat markets. This is just a visual check and it may just be my system or because I am only interested in trendy markets. When you say the 3 parameters I assume you mean length, fastend and slowend. I have only optimized on length. The problem IMO is that you then have three parameters to optimize on instead of just one for an exponential MA. I think the 'best' or most suitable selection of an MA must take this into account. Cheers
FWIW I did find that Adaptive MA's generate about 20% fewer trades than a Simple MA and about 10% less than Exponential and Weighted MAs. But that may be just my system.