1. Home
  2. Time series

Time series

Genstat provides several methods for examining and analysing time series. Sample correlation functions are produced by the directive CORRELATE:

    CORRELATE forms correlations between variates, autocorrelations of variates, and lagged cross-correlations between variates

The analysis of Box-Jenkins models is specified by several directives:

    FTSM forms preliminary estimates of parameters in time-series models
    TRANSFERFUNCTION specifies input series and transfer-function models for subsequent estimation of a model for an output series
    TFIT estimates parameters in Box-Jenkins models for time series (renamed version of ESTIMATE, which is retained as a synonym)

Information can be saved in Genstat data structures, or further output can be produced:

    TDISPLAY displays further output after an analysis by TFIT
    TKEEP saves results after an analysis by TFIT
    TFORECAST forecasts future values of a time series (renamed version of FORECAST, which is retained as a synonym)
    TSUMMARIZE displays characteristics of a time series model

It is also possible to filter a time series, or perform spectral analysis via the Fourier transform of a time series using the directives:

    TFILTER filters time series by time-series models (renamed version of FILTER, which is retained as a synonym)
    FOURIER calculates cosine or Fourier transforms of a real or complex series

Relevant procedures in the Library include:

    BJESTIMATE fits an ARIMA model, with forecasts and residual checks
    BJFORECAST plots forecasts of a time series using a previously fitted ARIMA
    BJIDENTIFY displays time series statistics useful for ARIMA model selection
    DFOURIER performs a harmonic analysis of a univariate time series
    KALMAN calculates estimates from the Kalman filter
    DKALMAN plots results from an analysis by KALMAN
    MCROSSPECTRUM performs a spectral analysis of a multiple time series
    MC1PSTATIONARY gives the stationary probabilities for a 1st-order Markov chain
    MOVINGAVERAGE calculates and plots the moving average of a time series
    PERIODTEST gives periodogram-based tests for white noise in time series
    PREWHITEN filters a time series before spectral analysis
    REPPERIODOGRAM gives periodogram-based analyses for replicated time series
    SMOOTHSPECTRUM forms smoothed spectrum estimates for univariate time series
    TVARMA fits a vector autoregressive moving average (VARMA) model
    TVFORECAST forecasts future values from a vector autoregressive moving average (VARMA) model
    TVGRAPH plots a vector autoregressive moving average (VARMA) model
Updated on May 20, 2019

Was this article helpful?