Premium
Estimating a Multivariate Familial Correlation Using Joint Models for Canonical Correlations: Application to Memory Score Analysis from Familial Hispanic Alzheimer's Disease Study
Author(s) -
Lee HyeSeung,
Cho Paik Myunghee,
Lee Joseph H.
Publication year - 2009
Publication title -
biometrics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.298
H-Index - 130
eISSN - 1541-0420
pISSN - 0006-341X
DOI - 10.1111/j.1541-0420.2008.01075.x
Subject(s) - canonical correlation , correlation , multivariate statistics , pairwise comparison , statistics , family aggregation , confounding , regression analysis , trait , regression , multivariate analysis , mathematics , econometrics , computer science , disease , medicine , geometry , pathology , programming language
Summary Analysis of multiple traits can provide additional information beyond analysis of a single trait, allowing better understanding of the underlying genetic mechanism of a common disease. To accommodate multiple traits in familial correlation analysis adjusting for confounders, we develop a regression model for canonical correlation parameters and propose joint modeling along with mean and scale parameters. The proposed method is more powerful than the regression method modeling pairwise correlations because it captures familial aggregation manifested in multiple traits through maximum canonical correlation.