ABSTRACT
Biostatistics is a science that develops statistical methodologies motivated by questions and scientific problems within the areas of medicine, epidemiology public health and biology. In this context, it is common to observe several individuals repeatedly over time for one or more responses of interest, generating then the so called longitudinal data. The main characteristics of these data are: measures from different subjects could be considered independent; but measures from a same subject at different time points should be considered, in some way, correlated. Therefore, specific longitudinal models have to be used in these data so that main assumptions are guaranteed. Mainly, these models allow us to distinguish, in the data, variability within and between subjects (Diggle et al, 2002) and describe the process under- neath the observed data. Different components of variability are considered, random effects and a stochastic process with a time correlation structure at individual level. In this seminar different longitudinal models are presented motivated by real data sets. Extended topics of longitudinal data analysis are presented, in particular, the issue of missing data and joint modelling of longitudinal and time-to-event data.
References
- [1] Diggle P, Heagerty P, Liang K-Y, Zeger S (2002), Analysis of Longitudinal Data, Oxford Statistical Science Series, 2nd series, Oxford University Press
- [2] Verbeke G, Molenberghs G (2009), Linear Mixed Models for Longitudinal Data, Springer Series in Statistics, Springer
- [3] Daniels M. J., Hogan J. W. (2008), Missing Data in Longitudinal Studies, Monographs on Statistics and Applied Probability, Chapman & Hall/CRC