Temporal data is information measured in the context of time. This contextualstructure provides components that need to be explored to understand the dataand that can form the basis of interactions applied to the plots. Inmultivariate time series we expect to see temporal dependence, long term andseasonal trends and cross-correlations. In longitudinal data we also expectwithin and between subject dependence. Time series and longitudinal data,although analyzed differently, are often plotted using similar displays. Weprovide a taxonomy of interactions on plots that can enable exploring temporalcomponents of these data types, and describe how to build these interactionsusing data transformations. Because temporal data is often accompanied othertypes of data we also describe how to link the temporal plots with otherdisplays of data. The ideas are conceptualized into a data pipeline fortemporal data, and implemented into the R package cranvas. This packageprovides many different types of interactive graphics that can be used togetherto explore data or diagnose a model fit.
展开▼