|
Activity Number:
|
108
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, July 30, 2007 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
IMS
|
| Abstract - #308937 |
|
Title:
|
Differentiated Logdensity Approximants
|
|
Author(s):
|
Serge Provost*+
|
|
Companies:
|
The University of Western Ontario
|
|
Address:
|
Dept of Stat and Act Sciences WSC 262, London, ON, N6A5B7, Canada
|
|
Keywords:
|
Density approximation ; Percentiles ; Moment-based methodologies ; Orthogonal polynomials ; Computational Statistics
|
|
Abstract:
|
An alternative density approximation methodology according to which the first derivative of the logarithm of the density approximant is expressed as a ratio of polynomials, is being proposed. The polynomial coefficients of the numerator and denominator of the rational function are determined by solving the system of linear equations which results from matching the moments of the approximant to those of the distribution being approximated. As will be explained, such approximants possess numerous properties of interest. The results will be compared in specific applications to those obtained from the saddlepoint technique and methodologies based on orthogonal polynomials and Edgeworth expansions. Several illustrative examples will be provided.
|
