Hello allstat -
Thanks to the dozen people who responded to my query about regression using
Excel. Here is a summary
1/
There were a couple of respondents who pointed out some weaknesses in the
intrinsic capabilities of Excel. Here is one reference:
* McCullough BD, Wilson B. 1999 On the accuracy of statistical
procedures in Microsoft Excel 97. Computational Statistics and Data Analysis
31:27-37.
2/
The following sites/comments have what appear to be useful Excel add-ins or
pointers to other resources. The packages appear/claim to be able to deal
with a large number of independent variables, and carry out their matrix
computations through extrinsic DLLs. Some offer free trial periods. I am
not familiar with the packages, so I pass them along with a 'caveat emptor'
* http://www.analyse-it.com/info/genmod.htm. - for the Analyse-It
extensions
* http://www.xlstat.com/ - the XLSTAT software
* http://www.statsdirect.com - the StatsDirect software
* http://www.nag.co.uk/stats/ae_soft.asp - the Numerical Algorithms
Group (NAg) software
* the KADD software available with "Data, Statistics, and Decision
Models with Excel" by Harnett, D.L. and Horrell, J.F., Wiley (1998)
* the UNISTAT software
3/
There is a list, ASSUME, devoted to statistics and Excel. See the mailbase
website for particulars:
http://www.stats.gla.ac.uk/allstat
Finally, for a list of software resources
http://www.stats.gla.ac.uk
Again, thanks to those who responded.
Donald C.E. Robinson
[log in to unmask]
ESSA Technologies Ltd
Vancouver, Canada
http://www.essa.com
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
