Subject: | |
From: | |
Reply To: | Classification, clustering, and phylogeny estimation |
Date: | Thu, 14 Nov 2002 17:23:04 -0500 |
Content-Type: | text/plain |
Parts/Attachments: |
|
|
Arthur:
You make astute and excellent points vis-a-vis using a proven statistical package. Having shared similiar experiences, your emphasize on the time to find source code for MDS, get function parameters correct is non-trivial and time consuming.
However the perspiration(!) from the effort of seeing the algorithim in action, how it's done, is invaluable to ascertain the accuracy of the results and fit of the model.
Last, the speed performance for hand written code is very demonstrable and significant.
I use as ancillary testing the proven packages to affirm the results of writting my own software.
In conclusion, I think it's balance of how large a data sample one has along with the virtue of really seeing the math and numerical methods behind the "blackbox". Your points are correct and the voice of experience!
Sincerely,
Neil Gottlieb
ProfileDepot, Inc.
wrote:
"...Using a package with proven track
record in human factors like SPSS can be very time-saving in the long
run. Use of syntax for one procedure easily generalizes to syntax for
others. Good quality assurance and documentation in the preparation
phase of analysis are crucial. Preparation of the data can easily be 80
or 90% of all the time in an analysis. With most software applications
you will have many attempts before getting the analysis done the way you
want it. With SPSS you will probably need fewer attempts. Even if you
use some other software for the final procedure, you would often find
SPSS very effective for the prep. Locating and downloading source code
can be time consuming. Learning how to adapt and compile source...Learning how to adapt and compile source
language code (FORTRAN, C++, IMSL) can take a lot of time. It also
means that you have to have access to them. It certainly took me much
more than a few hours to adopt SINDSCAL source code..."
|
|
|