[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
st: RE: discriminatory ability between typing systems
There are males, females and cyborgs of
indeterminate sex on this list, so the
address "Sir" is not advised.
Simpson's index has been programmed in
Stata as such in the program -ineq-
which is downloadable from SSC. That index
is based on summing the squares of
probabilities; sometimes people then
take the complement or the reciprocal
of that sum. Identical or similar indices
had been invented before, or were invented
after, Simpson published on this in 1949. He
himself got the idea from A.M. Turing
but could not say so because their contact was
work together on code-breaking in WWII.
Other names associated with this index
are those of Gini and Herfindahl, although
conversely the name of Gini has been
associated with several other measures.
I believe that geneticists refer to this
In short, the keyword to search under
is probably not Simpson, but inequality:
various versatile programs exist from
S. Jenkins, P. van Kerm and others.
This same Simpson is the person after whom
Simpson's paradox has been named, although
it was identified earlier by G.U. Yule
and even earlier by K. Pearson (and ...).
> How can I test the discriminatory ability between two
> bacterial genotypes typing systems ?
> I have heard about an application of the Simpson's index of
> diversity but
> didn't found anything in stata under this denomination. Is
> there anything to
> do with the hierarchical cluster analysis and what in this case?
* For searches and help try: