We will be posting a freeware software at www.yntisoftware.com for user
input on functionality [analysis, modeling, prediction, etc, etc]. The
software should be ready for download sometime in September [we're still
ironing around to not look to childish in math]. Just register your e-mail
there [no names] and you'll be notified when it is ready for download. The
bad news is that it is written for Windows and not Unix [one would assume
that there is more probability in finding access to a PC than to a Unix
box]. Anyways, check it out and make your choice [if after trying it you
would want it to have lagrange polynomial interpolation you can suggest so,
and so on -- so far I know it has fournier interpolation among others, all
kinds of linear fits [these include exponential, logarithmic, ARMA and
polynomial], some non-linear methods and free defined feedforward neural
networks with user defined activation functions, which can all be applied on
the same model to see the one you may prefer, extensive charting and
reporting, etc, etc.
<[EMAIL PROTECTED]> wrote in message 8lalui$655$[EMAIL PROTECTED]">news:8lalui$655$[EMAIL PROTECTED]...
> I am looking for way to automate numerical analysis. Specifically I
> want a program to take as input a time series from a flat text file,
> some settings on what analysis to perform, then use, for example,
> lagrange polynomial interpolation to produce a polynomial regression
> equation. From there I could write a program to graph predicited
> results. So far I've gone through applications, source libraries, and
> just today spoke with a sales rep from SAS for 30 minutes. However I
> may be making this more complex then it really is. Has anyone come
> across a similar problem or have advice on how to do this? SAS seems
> to be the only program that is flexible enough to run on unix and be
> automated. I would prefer a more simplistic approach as there is _a
> lot_ of overhead in starting to use SAS. Any help or suggestions would
> be appreciated!
>
> Thanks,
>
> -- Blaine
>
> [EMAIL PROTECTED]
>
> please CC my email address if you do reply :)
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
