Xu, Danli (Lois)
Area-proportional Visualisation for Circular Data
Danli (Lois) Xu
Department of Statistics, University of Auckland.
Data visualisation is important for statistical analysis, as it helps to communicate information efficiently and understand the significance of data in a visual context. It is particularly helpful to display circular data in a 2-dimensional space due to its non-linear support space and special characteristics, as they are able to reveal the underlying circular structure which is otherwise not obvious in 1-dimension.
In this talk, we will first formally categorise circular plots into two types, height- and area-proportional ones, and then describe a new general methodology to visualise circular data area-proportionally in a 2-dimensional space. Formulae are given that are fairly simple yet effective to produce desired circular histograms, rose diagrams and dot plots. More importantly, a density curve can now be produced and superimposed to match either the height- or area-proportional display of the raw data. This allows a visual assessment of the goodness of fit of selected models and enables a quick comparison among various models. In addition, plots are also developed for circular data with multiple classes.
This presentation is eligible for the NZSA Student Prize.