|
Abstract:
|
Longitudinal (panel) data provide the opportunity to examine temporal patterns of individuals, because measurements are collected on the same individuals at different, and often irregular, time points. One of the challenges with this data is the typical "spaghetti plot" visualisation, where a line plot is drawn for each individual, measured over time for some variable. Even a small number of individuals can make these plots too overplotted to parse, and interesting signals are lost to the noise. Statistical models can be fit to these data to understand them, but we can often miss the individual patterns.
This talk discusses new methods for identifying interesting individuals to better capture the individual experience. I introduce the R package, brolgar (BRowse over Longitudinal data Graphically and Analytically in R), which provides tools to facilitate these methods.
|
