1.1 Visualizing probability distributions across bivariate cyclic temporal granularities

Chapter 2 describes classes of time deconstructions using linear and cyclic time granularities, which can be used to create data visualizations to explore periodicities, associations, and anomalies. It provides a formal characterization of cyclic granularities and facilitates manipulation of single- and multiple-order-up time granularities through cyclic calendar algebra, as well as providing a recommendation algorithm to check the feasibility of creating plots for any two cyclic granularities. Our proposed method is also applicable to non-temporal hierarchical granularities with an underlying ordered index. The methods are implemented in the open-source R package gravitas and are consistent with a tidy workflow (Wickham and Grolemund 2016), with probability distributions examined using the range of graphics available in ggplot2(Wickham 2016).