Density Estimation from Aggregate Data

A kernel density estimator, constructed from a combination of disaggregate data subject to sampling bias and aggregate data, is described. The asymptotic performance of the estimator is explored, and details of an algorithm for its implementation are given. The issue of bandwidth selection is addres...

Full description

Saved in:
Bibliographic Details
Published inComputational statistics Vol. 19; no. 3; pp. 407 - 423
Main Author Hazelton, Martin
Format Journal Article
LanguageEnglish
Published 01.09.2004
Subjects
Aggregates
Asymptotic properties
Density
Estimators
Kernels
Sampling
Statistics
Vehicle detectors
Online AccessGet full text

Cover

More Information
Summary:A kernel density estimator, constructed from a combination of disaggregate data subject to sampling bias and aggregate data, is described. The asymptotic performance of the estimator is explored, and details of an algorithm for its implementation are given. The issue of bandwidth selection is addressed. Use of the estimator is illustrated through two examples. The first involves simulated data while the second example concerns traffic speed data collected by automatic vehicle detectors on Interstate 5 near Seattle.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:0943-4062
1613-9658
DOI:10.1007/BF03372104