Abstract Details
Activity Number:
|
30
|
Type:
|
Contributed
|
Date/Time:
|
Sunday, August 4, 2013 : 2:00 PM to 3:50 PM
|
Sponsor:
|
Biometrics Section
|
Abstract - #308247 |
Title:
|
Residuals in the Growth Curve Model and Their Application in the Analysis of Longitudinal Data
|
Author(s):
|
Jemila Hamid*+ and WeiLiang Huang
|
Companies:
|
McMaster University and McMaster University
|
Keywords:
|
Residuals ;
Growth Curve Model ;
Longtudinal Data ;
Decomposition ;
Model Diagnostics
|
Abstract:
|
Statistical models often rely on several assumptions - distribution assumption on outcome variables as well as relationship assumptions that link the outcome and independent variables. Modeling is not complete without performing model diagnostics. Residuals play important roles in assessing model fit, validating model assumptions as well as identifying outliers and/or influential observations. In univariate models, residuals are relatively simple to examine and they have been studied extensively. However, in multivariate models and in the analysis of longitudinal data, residuals have complex structure, are correlated and do not often follow the normal distribution. In this study, we consider decomposed residuals defined using the growth curve model. We also transformed these residuals to uncorrelated residuals using Cholesky decomposition of the covariance matrix. Extensive simulations demonstrate that the decomposed and transformed residuals perform well for assessing model fit, checking the normality assumption as well as identifying outliers. We illustrate performance of our method using real data set.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2013 program
|
2013 JSM Online Program Home
For information, contact jsm@amstat.org or phone (888) 231-3473.
If you have questions about the Continuing Education program, please contact the Education Department.
The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.
Copyright © American Statistical Association.