but in the old version you could ask
SAS how it sees the DDE stuff.
If Jay's advice doesn't work, open the Excel spreadsheet and highlight
the cells you want to bring into SAS dataset. Copy. In SAS, on one of
the menus
ute and I think a Bonferroni approach would work
okay here. There are various methods of computing the intervals, but
I would suggest a Wilson-type interval. (For a discussion, see the
2000 issue of the Journal of Statistical Software that discusses
setting CI's for the multinomial)
Warren M
I remember one reviewers comment was
something like "Why on earth would anyone want to use GEE's for such a
simple design". Hmm.
Warren
[EMAIL PROTECTED] (SR Millis) wrote in message
news:<[EMAIL PROTECTED]>...
> In November's issue of the American Statistician, Yan
the students, an OB-GYN
fellow, went to California last week and brought back a "sample" of
all blue M's from her meeting. It was great.
Warren May
=
Instructions for joining and leaving this list and remarks about
the p
I don't feel M&M's are any more trivial to introduce the concepts than
the usual "chips in an urn" or "balls from a jar". And I don't agree
that using M&Ms is trivial or juvenile. My class has basic science
grad students, M.D.s, Health professionals of all types and they do
seem to get something
the mean if it is "meaningless"?
Warren May
=
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=
I based on these samples. How would
you explain?
Thanks,
Warren
=
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=
Rafael,
Is this homeworks?
You're on the right track...you either need a separate "guess" for the
probability of success or you need an estimate from a sample.
Is there an even chance of turning right or left?
Are you a classicist, a frequentist or a subjectivist?
:)
Warren
[
I'm using Daniel's book too. I've used it for the last couple of
years, switching from Glantz Primer. The 7th edition still has quite
a few
errors, but I like it for some of the exercises.
Would love to be on your mailing list.
Warren
[EMAIL PROTECTED] (Robert Hamer) wrote i
http://bmj.com/cgi/content/full/322/7280/0
--
Warren S. Sarle SAS Institute Inc. The opinions expressed here
[EMAIL PROTECTED]SAS Campus Drive are mine and not necessarily
(919) 677-8000Cary, NC 27513, USA those of SAS Institute
e margin was less than 0.5%. And since the margin could be
either +0.5% or -0.5%, disputes would happen roughly twice as
often.
--
Warren S. Sarle SAS Institute Inc. The opinions expressed here
[EMAIL PROTECTED]SAS Campus Drive are mine and not necessarily
(919) 677-8000Cary
ly entered. But many new problems are also introduced. The
> opportunities for accidents and fraud are transformed into different
> categories -- such as tampering with software development and operation.
> And the desire for voter privac
?
The concepts of interval and ordinal scales are defined in terms
of the relationship between the scale and what it measures. You
can't know the scale type without understanding what it really
measures. See ftp://ftp.sas.com/pub/neural/measurement.html
--
Warren S. Sarle SAS Institut
rent images are not correlated unless you are doing
something weird. The paired t-test you propose is not invalidated
by spatial correlation. If you did McNemar's test on pixels, it
would not be valid due to spatial correlation.
Follow-ups set to sci.stat.consult
--
Warren S. Sarle
he major players have demos that
you might check. SPSS sent me a really nice SYSTAT CD with a bunch of
other Jandelly looking software like SigmaStat, etc. SPSS also has a
grad pack for a great price if you are a student. NCSS has a free
download that includes P
Textbooks still have the variance formulae in two forms...one for
computation. Why?
Warren May
(I'm still learning to teach, too)
Paul R Swank wrote:
> I found your syllabus to be very ambitious for undergraduates. Is this
> their first stat course?
>
> At 07:34 AM 6/18/00 -0400, S
if I
remember correctly. I think it might be like the old IMSL library...you
can't run the code without having ENTISOFT. Bummer.
www.entisoft.com
Luck,
Warren
Clay Fink wrote:
> I am reverse engineering a piece of software that was implemented in
> SAS. It uses the BETAINV function to c
class. Access to a computer with some REAL software is the
way to go. The $300 US you pay for a calculator will go a long way on
software (SPSS had a grad pack that was an awesome deal...hope they
still have it).
Warren May
Donald Macnaughton wrote:
> Stefanie Dzierzon writes (on 00/5/13)
&
therwise, don't test.
;)
Warren May
===
This list is open to everyone. Occasionally, less thoughtful
people send inappropriate messages. Please DO NOT COMPLAIN TO
THE POSTMASTER about these messages because the post
In article <007201bfb639$d40ee400$[EMAIL PROTECTED]>,
[EMAIL PROTECTED] (David A. Heiser) wrote:
>
> - Original Message -----
> From: Warren Sarle <[EMAIL PROTECTED]>
> To: <[EMAIL PROTECTED]>
> Sent: Thursday, May 04, 2000 12:23 PM
> Subject: Re: no cor
In article <002501bfb563$e8e6e8e0$[EMAIL PROTECTED]>,
[EMAIL PROTECTED] (David A. Heiser) wrote:
>
> - Original Message -
> From: Herman Rubin <[EMAIL PROTECTED]>
> To: <[EMAIL PROTECTED]>
> Sent: Wednesday, May 03, 2000 8:20 AM
> Subject: Re: no correlation assumption among X's in MLR
>
So, we must be forced to decide between two competing hypotheses?
H0: Use classical hypothesis testing
H1: Use Bayesian analysis
Is there enough evidence to reject the null?
Or, what is the probability that the null is true if we reject it?
Milo Schield wrote:
> Dennis Roberts has issued a fu
e enough is a matter for subject specialists.
Also, if I remember correctly you want to use 2 alpha instead of alpha for
constructing the CI.
Warren.
Rich Ulrich wrote:
> (posted to sci.stat.consult,sci.stat.edu, where version of the same
> post by Wang appeared.)
>
> On Tue, 28 Mar
to preach. Want to see what other
say.
Warren
dennis roberts wrote:
> for i don't know how many years now ... i have let my stat class students
> bring in a 3 by 5 notecard packed with whatever they want ... most put
> formulas ... thus, this is a legal cheat sheet. but, i have
significantly different from
zero is it? Your larger sample will allow you to estimate the mean with more
precision. (I'm not sure that I agree that larger samples are more problematic
from the terms of missing data than smaller samples...that's a problem with
study design, not sample size
re forced to analyze something without the aid of a
bookshelf full of reference materials?
I do feel plugging numbers into the formulaes is beneficial, but this might be
better handled with out of class assignments. Depends again on what you want to
test...computation or concept.
Warren
"T.S
y use some form of
multimedia...slides, overheads, etc. Afterwards, we open the floor for
discussion focusing on the statistical methods and results. Lots of great
opportunities (and sometimes exam questions) come from these discussions.
Anxious to hear how others feel about projects.
Warren
de
ate of the clinically important
difference(s). If so, why do you need a pilot study? If not, then you don't
have enough information to do a power study to begin with...well, you could use
Cohen's rather vague "small", "medium" and "large" criteria.
Warren May
ppreciate helping them make this
connection. Just presenting computational formulae for the "if you see
one of these, then do this test" isn't, IMHO, something students
remember for very long.
Anxious to see more of how people feel,
Warren May
>
> William J. Larson wrote:
&
nformation
about our statistical software and download evaluation versions
(http://www.kovcomp.com/).
--
Dr. Warren L. Kovach
Kovach Computing Services
85 Nant-y-Felin, Pentraeth, Anglesey
Wales LL75 8UY U.K.
Tel.: +44-1248-45
Thom Baguley wrote:
> christophe tourenq wrote:
> >
> > I am a biologist (definitely not a statistician) in France with very
> > limited access to statistical resources and I wonder if someone can give
> > me some advice. I have been conducting weekly bird counts in wetland
> > areas for a peri
Jerry Dallal wrote:
> dennis roberts wrote:
>
> > well, whether it is called a transcript or something else ... we
> > DO need some record of what the student did (i don't think having the
> > student say ... "I went to Purdue ... " would be sufficient)... and,
> > what courses a student t
a "quiz"
at the beginning of class by asking students to turn in a specific homework problem.
I usually pick one that is fairly far along in the homework set...students know this,
but they usually must do the "easy" ones in order to work up to the harder ones.
Sure, some do cheat but they usually fail the inclass tests anyway.
Warren.
33 matches
Mail list logo
