home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: sci.math.stat
- Path: sparky!uunet!sun-barr!ames!nsisrv!amarna.gsfc.nasa.gov!packer
- From: packer@amarna.gsfc.nasa.gov (Charles Packer)
- Subject: Re: Congressional Check Bouncing
- Message-ID: <10NOV199207082159@amarna.gsfc.nasa.gov>
- News-Software: VAX/VMS VNEWS 1.4-b1
- Sender: usenet@nsisrv.gsfc.nasa.gov (Usenet)
- Nntp-Posting-Host: amarna.gsfc.nasa.gov
- Organization: Dept. of Independence
- References: <DAVIS.92Nov8151949@pacific.mps.ohio-state.edu> <1992Nov9.225516.11794@Princeton.EDU>
- Date: Tue, 10 Nov 1992 11:08:00 GMT
- Lines: 12
-
- In article <1992Nov9.225516.11794@Princeton.EDU>, sjackman@phoenix.Princeton.EDU (Simon D. Jackman) writes...
- >distribution is highly positively skewed. I think it maxed out around
- >960 checks, but a large number of people had zero or only a small number
- >of bounced checks.
-
-
- ..And positive skewness is what one would expect, since check-bouncing is
- a performance variable, in the same class as income attainment or
- publishing papers! Experts, is this a lognormal or hyperbolic distribution?
- Does one shade into the other? In either case, aren't multiplicative
- effects usually proposed to model this, as opposed to the additive
- effects that generate a Gaussian distribution?
-