Panel data records multiple entities at multiple points in time, combining cross-sectional and temporal dimensions. Unlike cross-sectional data, it includes repeated observations; unlike ordinary time-series data, it follows multiple entities. section 17.1
A sales-and-purchases dataset illustrates the structure: multiple staff members make purchases at specific times. Irregular timestamps can be aggregated by entity and time period to create evenly spaced observations. section 17.1
From transactions to daily panels
Separate purchases and sales, index records by datetime, resample them by day, and combine the resulting columns for comparison over time.
