Hi again,
I appreciate the replies I've received so far. Here are a
few things I've learned so far from various texts and URLs.
QQ-plots comparing 2 empirical distributions:
1) If the quantile points lie along the (y=x) line, then
both distributions are similar in shape, spread and center.
2) If the points follow a straight line that is basically
parallel to y=x (however, slightly below or above it), the
distributions are similar in spread and shape but have
different centers.
3) If the points follow a straight line not parallel to
y=x, the distributions are similar in shape but have
different spreads. The center may also differ (but not
necessarily).
4) If the points do not follow a straight line, the
distributions have different shapes. Again, they MAY
differ in center and spread.
These seem to be good general guidelines, however, the qq-
plots I'm examining rarely seem to follow these patterns.
In fact, they often seem to be a combination of these
patterns. Two distributions may have qq-plots were the
first 50% of the data lie relatively close to y=x, then
there may be a sudden upwards/downwards curvature or a hump
(bell) like shape near the tail. I find it difficult to
interpret these types of situations other to say the
obvious, that there appears to be a difference in the right
tails of these two distributions.
Anyhow, back on research wagon.
Thanks again.
Patrick
[EMAIL PROTECTED]
* Sent from AltaVista http://www.altavista.com Where you can also find related Web
Pages, Images, Audios, Videos, News, and Shopping. Smart is Beautiful
