home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!usc!news.service.uci.edu!network.ucsd.edu!nprdc!craiger
- From: craiger@nprdc.navy.mil (J. Philip Craiger)
- Newsgroups: comp.apps.spreadsheets
- Subject: problems with linest() function
- Message-ID: <42991@arctic.nprdc.navy.mil>
- Date: 20 Jan 93 20:02:54 GMT
- Sender: news@nprdc.navy.mil
- Distribution: usa
- Organization: Navy Personnel R&D Center
- Lines: 22
-
-
- This solution should solve the problem, but it is difficult to believe
- that there isn't a more elegant solution to the problem. I also find
- the same problem when using any of the functions in the analysis
- toolpak of ver. 4.0. If i were to compute a 10x10 correlation matrix,
- i would have to go through an select out all cases with missing
- values. If Microsoft is to gain support from the scientific
- community, they should try to correct this problem. Very few real
- world data sets that I've come across are absolutely free of missing
- values. I think their analysis toolpak is great, but has serious
- limitations.
- --
- _______________________________________________
-
- J. Philip Craiger, Ph.D. (Internet: craiger@nprdc.navy.mil)
- Navy Personnel Research and Development Center -- Code 161
- San Diego, CA 92152-6800 (619-553-7961)
- --
- _______________________________________________
-
- J. Philip Craiger, Ph.D. (Internet: craiger@nprdc.navy.mil)
- Navy Personnel Research and Development Center -- Code 161
-