1.3 Missingness in time

Missing data always creates an obstacle in data analysis, because many techniques cannot operate without complete data. Another way to think about it is that the invisibles (missing values) may mask some data gems. The ability to explore missing value patterns without thought to later analyses is a worthwhile pursuit in itself. To support missing value handling for temporal data, a new data structure is designed which indexes missing values and efficiently stores the information. Building on this, several new operations and visualization methods have been designed. Chapter 4 describes these developments, and how they can be used to support exploration of missing patterns in time, and preparing data for imputation to feed into models. These methods expand the temporal data handling capabilities into a tidy workflow infrastructure.