Temporal Prediction System

A Temporal Prediction System is a predictive system that applies a forecasting algorithm to solve a forecasting task.

• AKA: Temporal Forecaster, Temporal Extrapolator.
• Context:
  • It can make use of a Forecasting Model.
  • It can range from being a Temporal Classification System to being a Temporal Ranking System to being a Temporal Estimation System.
  • It can range from being a Single-Predictor Forecasting System to being a Multi-Predictor Forecasting System.
  • It can range from being a Temporal Interpolation System to being a Temporal Extrapolation System.
  • It can be supported by a Forecasting Platform.
  • It can be supported by a Detrending System, such as an autocorrelation system.
  • It can range from being a Python-based Forecasting System, ...
  • ...
• Example(s):
  • a R forecast Package[1].
  • a Demand Prediction System.
  • an Event Duration Prediction System.
  • …
• Counter-Example(s):
  • a Noise-Filtering System.
• See: Classification System, Ranking System, Temporal Function, Monitoring System, Smoothing System.

---

---

References

2014

• https://algorithmia.com/algorithms/TimeSeries/Forecast
  • QUOTE: Gives a forecast the next n steps of a given time series based on extrapolation of linear and seasonal trends. It takes as input
    • time series as a double[].
    • number of time steps into the future to forecast.
    • maximum number of seasonal periods to consider.
  • Alternatively it takes just a time series, and defaults to a forecast range equal to the length of the original series and the single strongest seasonal trend.

    It returns the forecasted series as a double[]. This algorithm works by fitting a linear trend to the given data, extrapolating it into the future series interval, and then adjusting it based on the expected contributions of each detected seasonal component. Seasonality is detected using https://algorithmia.com/algorithms/TimeSeries/Autocorrelate.

2013

• http://www.businessdictionary.com/definition/forecasting-system.html
  • QUOTE: Set of techniques or tools required for analysis of historical data, selection of most appropriate modeling structure, model validation, development of forecasts, and monitoring and adjustment of forecasts, etc.

2002

• (Leigh et al., 2002) ⇒ William Leigh, Russell Purvis, and James M. Ragusa. (2002). “Forecasting the NYSE Composite Index with Technical Analysis, Pattern Recognizer, Neural Network, and Genetic Algorithm: A Case Study in Romantic Decision Support.” In: Decision Support Systems, 32(4).
  • QUOTE: ...and (3) approaches recently developed that combine diverse classification and forecasting systems. ...

---