Title: Polynomial Trendlines

Rakesh,

 

As stated in Appendix 6 of Hurst’s book and the “me too” but more fully fleshed out math examples in Claud Cleeton’s book ( Out of Print but still available used etc ) …

 

One can supposedly remove the trend from a CMA by subtracting a 1st order least square fit of the CMA from it which should leave a sinusoid of sorts.

 

This then should be able to be passed through a least squares trig fit to arrive at a frequency ( wave length ), amplitude and phase which can then be extrapolated.  The principle is the same as an FFT but utilizes shorter ( relative to the wave length being investigated ) data segments, which is probably a good idea since these things have a tendency to change character.

 

I have played with this somewhat ( See attached ) …

 

The top pane contains a CMA like MA and you can see where its last calculation is relative to the Selected Bar i.e. several bars behind.  It also has surrounding bands … In this case percentage based …

 

The green line going through the CMA is a least squares fit of the CMA over roughly one cycle.

 

The white line in the bottom pane is the original CMA with the least squares fit from above subtracted from it …

 

The alternating green and red sinusoid is the least squares trig fit of the white line and has been extrapolated into the future …

 

Although probably not really necessary except from an edge band perspective, This then should be able to be added back to the trend to extrapolate the CMA and its associated bands.

 

One can see that at times this is not a horrible predictor … However, the predictions have a tendency to be somewhat “twitchy” …

 

More study is needed to see if there is anything really usable here …

 

 

 

 



I am using the free version of SPAMfighter for private users.
It has removed 8605 spam emails to date.
Paying users do not have this message in their emails.
Try SPAMfighter for free now!
__._,_.___

Please note that this group is for discussion between users only.

To get support from AmiBroker please send an e-mail directly to
SUPPORT {at} amibroker.com

For other support material please check also:
http://www.amibroker.com/support.html






SPONSORED LINKS
Software support Small business finance Business finance online
Business finance training Business finance course

Your email settings: Individual Email|Traditional
Change settings via the Web (Yahoo! ID required)
Change settings via email: Switch delivery to Daily Digest | Switch to Fully Featured
Visit Your Group | Yahoo! Groups Terms of Use | Unsubscribe

__,_._,___

Attachment: CMA - 1st Order Detrend - Trig Fit.png
Description: Binary data

Reply via email to