Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: st: Concordance
From
"Nick Cox" <[email protected]>
To
<[email protected]>
Subject
RE: st: Concordance
Date
Sun, 14 Mar 2010 17:59:22 -0000
Thanks to Tom for mentioning -concord-, which is by Thomas Steichen and
myself.
In fact, a more recent version of -concord- than that on SSC is
downloadable from the Stata Journal website, as -search concord-
indicates.
If Ricardo had used -search concordance- he would got to this advice
from Tom Trikalinos and myself in a matter of seconds.
The Statalist FAQ mentions the use of -search- as a way to solve your
problems, as do many introductions to Stata.
Nick
[email protected]
Tom Trikalinos
i'm off to a meeting -- thus not time for a proper response, but these
are pointers.
1. difference vs average plots
2. Lin's concoradance correlation coefficient rather than Pearson
correlation
3. (reduced) major axis regression and related regressions ([weighted
] Deming, Passing-Bablock[spelling may be off?]) rather than OLS>
ssc install concord
will get you quite far actually.
N. orsini had written a deming routine i f I racall correctly.
On Fri, Mar 12, 2010 at 1:57 PM, Ricardo Ovaldia <[email protected]>
wrote:
> Glucose on 100 serum samples was measured by two different methods. I
would like to examine the concordance between the two set of
measurements. The glucose values are continuous, so I was thinking about
simply computing Pearson's correlation coefficient, but that seems too
simple and the test is not very sensitive. Is there a better way to
assess concordance between two repeated measurements.
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/