[S] calculation of probabilities from an empirical probability density function

hatfield (hatfield@westland.com)
Tue, 22 Sep 1998 22:53:36 -0400


Dear S+ users;

I am a biologist trying to quantify "uncertainty" in a statistical
relationship. A regression line between the two variables is highly
significant, but explains only about 70% of the variance. N is large, so a
simultaneous confidence limit around the regression line indicates a high
degree of confidence in the placement of the line, but does not properly
represent the scatter. I've used a density plot to show the distribution of
the residuals. Now I would like to calculate a probability from this
function, but I am unsure how to calculate the area under such a
non-parametric empirical density function. I'd like to be able to say that
some new point has x probability/likelihood of falling within y (upper) and
z (lower) boundaries. Any help is much appreciated.

(I am using S+ 3.3 for Windows)

Todd Hatfield

-----------------------------------------------------------------------
This message was distributed by s-news@wubios.wustl.edu. To unsubscribe
send e-mail to s-news-request@wubios.wustl.edu with the BODY of the
message: unsubscribe s-news