Premium
RANDOM SAMPLING IN PATTERN RECONSTRUCTION
Author(s) -
Moore Marc
Publication year - 1980
Publication title -
australian journal of statistics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.434
H-Index - 41
eISSN - 1467-842X
pISSN - 0004-9581
DOI - 10.1111/j.1467-842x.1980.tb01183.x
Subject(s) - realization (probability) , partition (number theory) , set (abstract data type) , simple random sample , computer science , sampling (signal processing) , mathematics , statistics , stratified sampling , sample (material) , random function , algorithm , random variable , combinatorics , computer vision , physics , population , demography , filter (signal processing) , sociology , thermodynamics , programming language
Summary A random pattern is the realization of a random set process giving a partition of a set. We are interested by the situation where an unknown random pattern has to be estimated (reconstructed) from sample data. Two sampling strategies to do so are presented. One is of random character the other is not. The risk corresponding to each is obtained. Some numerical comparisons are made for a particular model.