Date:         Wed, 1 May 1996 18:59:46 -0500
Reply-To:     GERALD ZUCKIER <ZUCKIER@CHIME.ORG>
Sender:       "SAS(r) Discussion" <SAS-L@UGA.CC.UGA.EDU>
From:         GERALD ZUCKIER <ZUCKIER@CHIME.ORG>
Subject:      Re: Exponentiation of Negative Numbers
In-Reply-To:  <199605011547.PAA29781@mag-net.co.uk>

But that sort of gets an arbitrary root. As pointed out, a kth root has k
solutions, just as (1)^.5 = 1 or -1; for square roots of positive numbers we
routinely use the positive one, although forgetting about the other one
sometimes gets us into trouble. Might picking the real root of a negative number
where such exists (odd roots) be just too arbitrary? (Not a rhetorical
question.) On the other hand, how hard would it be for SAS Institute to code up
a complex arithmetic module?