Sliding simulation: A new approach to time series forecasting

    Research output: Contribution to journalArticlepeer-review

    20 Citations (Scopus)

    Abstract

    This paper proposes a new approach to time series forecasting based upon three premises. First, a model is selected not by bow well it fits historical data but on its ability to accurately predict out-of-sample actual data. Second, a model/method is selected among several run in parallel using out-of-sample information. Third, models/methods are optimized for eacb forecasting horizon separately, making it possible to have different models/methods to predict each of tbe m horizons. Tbis approach outperforms the best method of the Af-Competition by a large margin wben tested empirically with the 111 series subsample of the Af-Competition data.

    Original languageEnglish
    Pages (from-to)505-512
    Number of pages8
    JournalManagement Science
    Volume36
    Issue number4
    Publication statusPublished - 1 Apr 1990

    Keywords

    • Accuracy Measures
    • Forecasting
    • M-Competition
    • Sliding Simulation
    • Time Series

    Fingerprint

    Dive into the research topics of 'Sliding simulation: A new approach to time series forecasting'. Together they form a unique fingerprint.

    Cite this