Simulating the Continuation of a Time Series in R

Abstract

The simulation of the continuation of a given time series is useful for many practical applications. But no standard procedure for this task is suggested in the literature. It is therefore demonstrated how to use the seasonal ARIMA process to simulate the continuation of an observed time series. The R-code presented uses well-known modeling procedures for ARIMA models and conditional simulation of a SARIMA model with known parameters. A small example demonstrates the correctness and practical relevance of the new idea.