A Mixture Multivariate Approach for Predicting Cardiovascular Health

• Activity Number: 44
• Type: Contributed
• Date/Time: Sunday, July 31, 2011 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistics in Epidemiology
• Author(s): Tamika Royal-Thomas*+ and Dan McGee and Debajyoti Sinha and Clive Osmond and Terrence Forrester
• Companies: Florida State University and Florida State University and Florida State University and University of Southampton and University of the West Indies
• Keywords: Principal component analysis ; linear mixed model ; longitudinal data ; Cardiovascular ; Multivariate approach

Abstract

Predicting coronary heart disease (CHD) has been researched extensively and there are still areas to uncover. Some of these areas include what happens during the early stages of human development which occurs inside the womb in pregnancy (in utero) and in early childhood. One of the difficulties with in utero and early childhood data is that certain variables are highly correlated and so using dimension reduction techniques are quite applicable in this scenario. Principal component analysis (PCA) is utilized in creating a smaller dimension of uncorrelated data which is then utilized in a longitudinal analysis setting. This work examines a unique longitudinal data which utilizes a mixture of multivariate approaches such as PCA in obtaining composite scores in utero and at birth of babies. These composite scores are then utilized in an optimal linear mixed model for longitudinal data which indicates that in utero and early childhood attributes predicts the future cardiovascular health of the children.