Adaptive MAs

06:25 @Ensign_H: After showing DYO stuff like this example, there will be no end to the research possibilities. The idea behind this example is I wanted a moving average that would adapt to the type of market. When the market is moving strongly up or down, I wanted the average to zoom in close to the data, and when the market is meandering sideways, I wanted the average to back off so as to not get caught in the chop.

06:28 @Ensign_H: This example does a pretty good job of doing that by adjusting the average parameter between 30 and 1.

06:29 @Ensign_H: A 5 bar Momentum is used as the barometer to adjust the parameter. When Momentum (Line B) is near zero the parameter calculated by lines C, D, and E will be 30. When Momentum is steep, say +5 pts, then the parameter calculated will be 30 - 5 * 5 = 5, thus the steeper the momentum the smaller the Average parameter will be.

06:32 @Ensign_H: Internally we check the average parameter and prevent it from being below 1...... therefore the formula is safe. If Momentum were to be higher than 6 or -6, the average will use a parameter of 1 because of internal parameter checking for our averages.

06:34 @Ensign_H: The example shows how the average zoomed in to keep pace with the sharp sell off after 9 am on the 17th, and then tracked more gently staying away from the data a bit as the market chopped higher from 9:30 to noon on the 17th, and chopped lower from noon on the 17th through the end of the chart example.

06:36 @Ensign_H: Other averages would have got caught in the chop. Let me show normal averages, one with a small parameter and one with a big parameter.

06:38 @Ensign_H: In this example 2 smoothed averages have been added, the Blue at parameter 20, and the Red at parameter 5. As can be seen, the 20 Blue tracks closely our Adaptive average on the left side of the chart and on the right side of the chart. However, the Blue really lagged behind the market in the sharp sell off after 9:00 on the 17th.

06:40 @Ensign_H: If we try to have a small parameter of a 5 Red to act faster to the sell off, then the Red line is caught in the market chop everywhere else.

06:41 @Ensign_H: So the Adaptive Average in Magenta on the chart is picking a bit of the best of both types of average. Enjoy your investigation of this type of adapting parameter.

06:46 @Ensign_H: One possible way to have a signal is to put on an average of the adaptive average and use the crossing of the two as the direction.

06:47 @Ensign_H: When adaptive is above its average, trade the long direction, and when it is below the average, trade the short side.

06:52 @Ensign_H: Here is an addition to the DYO to now do bar coloring for the direction of the trade from the average crossovers.....looks promising.

06:53 @Ensign_H: Anyway, these ideas are to teach you skills in using the DYO.

Thanks Howard! that's really great! will play with it!

06:55 @Ensign_H: welcome. bye

Last updated 02/22/2005