Kbg. Dear et al., ESTIMATING CORRELATIONS FROM EPIDEMIOLOGIC DATA IN THE PRESENCE OF MEASUREMENT ERROR, Statistics in medicine, 16(19), 1997, pp. 2177-2189
Analysis of a major multi-site epidemiologic study of heart disease ha
s required estimation of the pairwise correlation of several measureme
nts across sub-populations. Because the measurements from each sub-pop
ulation were subject to sampling variability, the Pearson product mome
nt estimator of these correlations produces biased estimates. This pap
er proposes a model that takes into account within and between sub-pop
ulation variation, provides algorithms for obtaining maximum likelihoo
d estimates of these correlations and discusses several approaches for
obtaining interval estimates. (C) 1997 by John Wiley & Sons, Ltd.