Premium
Sampling a finite population in the presence of trend and correlation: Estimation of total 305‐day lactation production in cattle
Author(s) -
Bartlett Roy F.
Publication year - 1986
Publication title -
canadian journal of statistics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.804
H-Index - 51
eISSN - 1708-945X
pISSN - 0319-5724
DOI - 10.2307/3314797
Subject(s) - estimator , statistics , mathematics , estimation , best linear unbiased prediction , population , covariance , milk production , dairy cattle , sampling (signal processing) , production (economics) , econometrics , zoology , demography , biology , selection (genetic algorithm) , computer science , economics , management , macroeconomics , filter (signal processing) , artificial intelligence , sociology , computer vision
This paper studies the estimation of a finite population total in the presence of trend. A practical problem of dairy science is to estimate a cow's total 305‐day milk production given a number of test‐day records. We analyze this problem as one of estimating the total of a discrete population when the population values are correlated and exhibit a trend over time. Linear prediction estimators that are BLUE for known covariance and trend function linear in unknown parameters were applied to the estimation of the milk yield total. An empirical study compares BLUE with the expansion estimator and the procedure currently used by the Canadian Record of Performance for Dairy Cattle.