With regards to your questions ...

I've played with a variety of methods ... 

At the moment the method I like is to ... 

- Normalize the data i.e. nData = Log10(Data)
- Calculate detrended data as nData - a first order LR of the nData
- Calc a x-bar CMA of detrended data
- Take a y-bar sampling of the CMA
- Calc a cubic spline to fill the gaps
- Trig fit the resulting cubic spline

There are a variety of other ways to play with this ... Some of which 
I have probably overlooked.


Reply via email to