2.7 Discussion

Exploratory data analysis involves many iterations of finding and summarizing patterns. With temporal data available at ever finer scales, exploring periodicity can become overwhelming with so many possible granularities to explore. This work provides tools to classify and compute possible cyclic granularities from an ordered (usually temporal) index. We also provide a framework to systematically explore the distribution of a univariate variable conditional on two cyclic time granularities using visualizations based on the synergy and levels of the cyclic granularities.

The gravitas package provides very general tools to compute and manipulate cyclic granularities, and to generate plots displaying distributions conditional on those granularities.

A missing piece in the package gravitas is the computation of cyclic aperiodic granularities which would require computing aperiodic linear granularities first. A few R packages including almanac (Vaughan 2020) and gs (Laird-Smith 2020) provide the tools to create recurring aperiodic events. These functions can be used with the gravitas package to accommodate aperiodic cyclic granularities.

We propose producing plots based on pairs of cyclic granularities that form harmonies rather than clashes or near-clashes. A future direction of work could be to further refine the selection of appropriate pairs of granularities by identifying those for which the differences between the displayed distributions is greatest and rating these selected harmony pairs in order of importance for exploration.