Seminar
Notice
Statistical
Laboratory
Iowa State University
DATE AND
TIME: Tuesday, June 27 2:00
p.m.
PLACE:
Sweeney 1116
SPEAKER:
Norma
Leyva-Estrada
TITLE: Approximate likelihood-based inference for weight-size
distributions
from sieving data
ABSTRACT
In sieving there is a fixed number of sieves
with progressively smaller size
openings. To study the size distribution of
solid particles using sieving, the
weight (not number) of the particles in
each size interval is obtained. This
produces a sample weight-size
distribution. We assume that a sample of particles
can be described as
randomly selected from a large population until a target
total weight is
reached. Also, we assume that the size follows a standard
probability density
(lognormal) and that the first and second moments of the
conditional
distribution of weight given size can be described with a power
law
relationship. Under these assumptions, we obtain approximate maximum
likelihood
point and interval estimates of the population weight fractions,
based on an
asymptotic multivariate normal distribution. We study the
properties of these
estimators via simulation. Standard errors of the maximum
likelihood estimates
of weight-size fractions are smaller than those of
estimators previously
proposed in the literature, and the confidence
intervals hold their nominal
coverage
value.
COFFEE: 1:45 a.m., 104
Snedecor Hall
Seminar schedules and abstracts are available via
WWW: http://www.stat.iastate.edu/
Denise Riker, Undergraduate Secretary
Department of
Statistics
Iowa State University
320 Snedecor Hall
Ames, IA
50011
(515) 294-1076