Time Series

Version 10 introduces systemwide support for time series, whether they originate in finance, economics, signal processing, fitness, or other areas. Reading data from connected devices and accessing knowledge from data paclets are rich sources of time series data. Data analytics on time series, including descriptive statistics, visualization, or automated model fitting, etc., now just works.

  • Construct time series starting from real-world data. »
  • Represent discrete-time data as event series. »
  • Visualize and compare different time series. »
  • Work with regular or irregularly sampled time series.
  • Use dates in time series as strings, lists, or date objects, supporting time zones, different calendars, or holiday schedules.
  • Work with quantities in time series.
  • Smooth a time series using moving averages or other filters.
  • Aggregate values of time series.
  • Map an arbitrary function over values of time series. »
  • Add, multiply, or otherwise combine time series.
  • Extract a fixed window from a time series. »
  • Compute descriptive statistical measures.
  • Capture time series from connected devices and external services.
  • Analyze time series from astronomy, finance, demographics, and other areas.
  • Generate parametric models that fit a time series.