Lee, and everybody: Oh, you have done me a wonderful boon here. I had no
idea that this PAST software even existed. As everybody probably already
knows, I'm not a big fan of biometry or biostatistics, and that's on
philosophical grounds; but I also confess to not being terribly good at
it. It isn't that I can't understand the mathematical concepts -- the big
problem, and this has gone on since I was in college, back in the day when
you took a stack of punchcards from one end of a big building at U. Mich.
and turned 'em in a block away at the other end -- is simply getting the
software to WORK. IT NEVER WORKS!!!!!!!!
BUT....I bravely downloaded the back-version of PAST that you mentioned,
and started playing around with it, and lo and behold, almost unbelievable
to me, it's not only straightforward to use, or what people call
"intuitive"; it also isn't terribly quirky, i.e. that you have to do this
or that or know some little trick before it will actually perform the
function you have requested. What happened instead was that I got a
perfectly usable cluster analysis the first time I asked! And then some
nice log-log bivariate plots! And then even a principal components
analysis on a big 8D matrix!
So I am overjoyed!
BUT....there is one problem, and maybe you or somebody else who is
experienced with PAST can help me and it won't be too much trouble. I
would really like to do a discriminant function analysis, but when I
highlight the columns to be included in the analysis, I get an error
message that says "two columns not colored correctly." Even when I use
"select all" it gives me this message. So this is the usual quirk that
lurks at the heart of all software, I guess....
Would someone be willing to send me the steps that I have to take to get a
D.F./Hotelling analysis on 8 or 9 parameters, with about 100 instances?
Extremely grateful....Dr. Deb
> Dear all,
>
>
>
> I just thought I'd mention that a new version of the PAST statistics
> package
> was released in November in case it was of interest to list members. This
> appears to be a substantial update and I know many other zooarchaeologists
> aside from myself used the previous version.
>
>
>
> The new version is still free and can be found here:
> http://folk.uio.no/ohammer/past/
>
>
>
>
>
> Best regards,
>
>
>
> Lee G. Broderick. BA (Hons), MSc, FZS
>
> Zooarchaeologist
>
> www.zooarchaeology.co.uk
>
>
>
> This e-mail is intended only for the person or entity to which it is
> addressed and may contain information that is private and confidential. If
> an addressing or transmission has misdirected this e-mail, please notify
> the
> author by replying to this e-mail. If you are not the intended recipient
> you
> must not use, disclose, copy, print or rely on this e-mail.
>
>
>
> We sweep all outgoing messages for the presence of computer viruses.
>
> However, we cannot accept any responsibility for any loss or damage to
> your
> systems due to viruses or malicious code not detected.
>
>
>
>
>
>
|