 |
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
Re: st: confidence interval for median
Also--
Feinstein did not state that the jackknifed standard error for the
median was accurate. He stated, according to your quote, only that
leave-one-out estimates of the median are easy to compute by hand
(true) and that they can, therefore, serve as "an excellent screening
test for stability of the median". This is nonsense. The median is
stable unless observations to its left are clearly separated from
observations to its right. A dotplot, histogram, or stem and leaf
plot would quickly show if this were the case.
-Steve
On Aug 10, 2008, at 2:22 AM, Maarten buis wrote:
--- moleps islon <[email protected]> wrote:
You state that jackknife is inappropriate. Principles of medical
statistics, feinstein 2002 claim the contrary-I have no clue what is
right..
I learned that from Efron, B. and Tibshirani, R. (1993). An
Introduction to the Bootstrap. Chapman & Hall/CRC. The logic is that
the statistic needs to be a smooth function of the data in order for
the jackknife to work, and the median obviously is not a smooth
function of the data.
-- Maarten
-----------------------------------------
Maarten L. Buis
Department of Social Research Methodology
Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands
visiting address:
Buitenveldertselaan 3 (Metropolitan), room Z434
+31 20 5986715
http://home.fsw.vu.nl/m.buis/
-----------------------------------------
__________________________________________________________
Not happy with your email address?.
Get the one you really want - millions of new email addresses
available now at Yahoo! http://uk.docs.yahoo.com/ymail/new.html
*
* 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/
*
* 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/