Move the moving average?
| |
|
Last Poster |
Posts |
Pages |
Last Post |
| |
|
|
|
|
|
|
| Author |
Message |
   engcomp
Member
Username: engcomp Post Number: 1 Registered: 08-2007Rating: N/A Votes: 0
|
| | Thursday, August 02, 2007 - 07:59 am: | 
|
Is it possible somehow to move the moving average forward or backward relative to its current position? For example, in one trading system, the MA is moved forward by the square root of the period, which better simulates trend lines. In another system, very successful for me, the MA is moved back by half its period (i.e. it is centered) and then spaced vertically by a certain size, to create a curved trend channel (see http://futuresandforex.info/schm.shtml).
|
   colin_twiggs
Member
Username: colin_twiggs Post Number: 2999 Registered: 09-2002Rating: N/A Votes: 0
|
| | Tuesday, August 07, 2007 - 11:27 am: | 
|
engcomp, Thank you for the suggestions. We plan to upgrade the indicator selection when we have completed the drawing tools upgrade, later this year. I have added your request to our list for investigation. Regards, Colin
|
   engcomp
Member
Username: engcomp Post Number: 2 Registered: 08-2007Rating: N/A Votes: 0
|
| | Tuesday, August 07, 2007 - 03:02 pm: | 
|
Hi, Colin The following also went by email, but not sure if it was delivered: Thank you for your interest in these suggestions. If you need a volunteer to assist, please feel free to call on me. Repositioning the MA by a given +/- period should be easy. The SCHM bands are slightly more difficult: 1. To find the harmonic mean, you must first reshape the n data items x into 1/x, then find their arithmetic mean (am), then inverse the answer to get MA = 1/am. The effort is worth doing because the result is a much smoother line. 2. The endpoint of the line is centered on the data, therefore, n should be an odd number. The vertical spacing for the top and bottom edges of the bands can be done by eye if the line can be grabbed and copied (not a trivial challenge). 3. At least two bands are required - an inner band that contains the data x, and an outer band that contains the inner band. Ideally, the outer band is further contained within a regression channel. Please keep me posted of your progress, Helmut
|
   colin_twiggs
Member
Username: colin_twiggs Post Number: 3000 Registered: 09-2002Rating: N/A Votes: 0
|
| | Tuesday, August 07, 2007 - 03:19 pm: | 
|
Thank you Helmut, We might need your assistance when we tackle SCHM bands. Regards, Colin
|
|
Threads by Last Post Time: First
Previous
68
69
70
71
72
73
74
75
76
77
Next
Last
|