home *** CD-ROM | disk | FTP | other *** search
/ NetNews Usenet Archive 1992 #27 / NN_1992_27.iso / spool / bit / listserv / statl / 2003 < prev    next >
Encoding:
Text File  |  1992-11-17  |  1.7 KB  |  30 lines
  1. Comments: Gated by NETNEWS@AUVM.AMERICAN.EDU
  2. Path: sparky!uunet!stanford.edu!bcm!convex!news.utdallas.edu!wupost!gumby!yale!yale.edu!news.yale.edu!mars.caps.maine.edu!maine.maine.edu!cunyvm!psuvm!auvm!VTVM1.BITNET!WLTPIRIE
  3. Message-ID: <STAT-L%92111714334241@VM1.MCGILL.CA>
  4. Newsgroups: bit.listserv.stat-l
  5. Date:         Tue, 17 Nov 1992 14:09:04 EST
  6. Sender:       "STATISTICAL CONSULTING" <STAT-L@MCGILL1.BITNET>
  7. From:         Walt Pirie <WLTPIRIE@VTVM1.BITNET>
  8. Subject:      Transformations
  9. Lines: 19
  10.  
  11. No criticism intended, but I have never understood the concern about "natural s
  12. cale or metric" when using transformations. If the transformation is (hopefully
  13. ) 1 to 1, why is that a concern? And why is it any less "natural" than using a
  14. "standardization" transformation to the t-test scale?
  15. Not that I like transformations, but for other reasons. Like what they do to er
  16. ror structure and related assumptions.
  17.  
  18. |==================================|============================|
  19. |  Walter R. Pirie                 |                            |
  20. |  Department of Statistics        |                            |
  21. |  Virginia Tech                   |                            |
  22. |  Blacksburg, VA 24061-0439       |                            |
  23. |                                  |                            |
  24. |  Tel. 703-231-5441               |                            |
  25. |  Bitnet WLTPIRIE@VTFVM1          |                            |
  26. |  Telnet WLTPIRIE@VTVM1.CC.VT.EDU |                            |
  27. |                                  |                            |
  28. |  Fax  703-231-3863               |                            |
  29. |==================================|============================|
  30.