Visualizing Probability Distributions Across Bivariate Cyclic Temporal Granularities

Deconstructing a time index into granularities can assist in exploration and automated analysis of large temporal datasets. This article describes classes deconstructions using linear cyclic granularities. Linear respect the progression such as hours, days, weeks months. Cyclic be circular hour-of-the-day, quasi-circular day-of-the-month, aperiodic public holidays. The hierarchical structure creates nested ordering: hour-of-the-day second-of-the-minute are single-order-up. Hour-of-the-week is multiple-order-up, because it passes over day-of-the-week. Methods provided for creating all possible index. A recommendation algorithm provides an indication whether pair meaningfully examined together (a “harmony”), or when they cannot “clash”). Time used to create data visualizations explore periodicities, associations anomalies. form categorical variables (ordered unordered) which induce groupings observations. Assuming numeric response variable, resulting graphics then displays distributions compared across combinations variables. methods implemented open source R package gravitas consistent with tidy workflow, probability range available ggplot2. Supplementary files this online.