Premium
Non‐random sampling in human genetics: Estimation of familial correlations, model testing, and interpretation
Author(s) -
Chakraborty Ranajit,
Hanis Craig L.
Publication year - 1987
Publication title -
statistics in medicine
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.996
H-Index - 183
eISSN - 1097-0258
pISSN - 0277-6715
DOI - 10.1002/sim.4780060511
Subject(s) - randomness , interpretation (philosophy) , context (archaeology) , estimation , statistics , sampling (signal processing) , econometrics , population , correlation , computer science , mathematics , biology , demography , paleontology , management , geometry , filter (signal processing) , sociology , economics , computer vision , programming language
Abstract By choice or necessity, human geneticists and genetic epidemiologists often design studies that involve non‐random sampling of clusters of individuals, and yet address hypotheses appropriate to the population as a whole. Failure to adjust for the non‐randomness of data often leads to biased parameter estimates and misspecification of predictive models that cause familial resemblance of traits. We develop an approach to adjust for common forms of non‐randomness in the context of estimating familial correlation with minimal distributional assumptions and discuss its implications in connection with adjustments for concomitant variables.