Topic:  Longitudinal Analysis of Multivariate Responses.

Date:  Monday 25^th October 13h00-14h00

Zoom link:  https://uct-za.zoom.us/j/97481774461?pwd=MlQ4T1IzbCtiUksrbHBVT2F4MnV1Zz09
Meeting ID: 974 8177 4461
Passcode: 376775

The Dean of the Faculty of Science, Professor Maano Ramutsindela, invites you to the inaugural lecture of Professor Francesca Little from the Department of Statistical Sciences.  Professor Little will deliver her inaugural lecture entitled, "Longitudinal Analysis of Multivariate Responses", on Monday 25^th October at 13h00-14h00.

There is much awareness these days about the availability of lots of data and methods for dealing with that. In the medical sciences, one of the ways in which lots of data are generated is through the repeated measurement of outcomes of interest at multiple successive timepoints. Another is the measurement of many different but related outcomes for the same subjects. The presentation will focus on the statistical modelling of such complex data from the medical sciences.

Longitudinal Data Analysis refers to the modelling of repeated measures of the same response for subjects over time. There are various approaches to model these data that take into account the within-subject correlation, the most popular being the use of mixed effect models. Increasingly, research studies do not focus on just one longitudinal response over time, but multiple such responses. For example:

1. Different cytokine measurements as possible immune markers for TB.
2. Different brain metabolite measurements in young children.

Prof Little's research activities originate from her collaboration with the Health Sciences, predominantly in the areas of Malaria, TB and HIV research and the analysis of data from longitudinal birth cohorts, leading to an interest in statistical methodological topics of nonlinear mixed effect modelling, zero-inflated mixture distributions for discrete data, analysis of time to event outcomes, casual modelling in observational studies, longitudinal studies with missing data, growth curve modelling, simulation models for infectious diseases, latent variable modelling and multivariate analysis. Her main contribution to these research collaborations is the use of state-of-the-art statistical modelling to enhance the quality of the medical research, in order to make a difference to the quality of life of vulnerable populations in Southern Africa.