Graphical Tools for Exploring and Analyzing Data From ARIMA Time Series Models
William Q. Meeker · Iowa State University Digital Repository (Iowa State University) · 1999
S-plus is a highly interactive programming environment for data analysis and graphics. The S-plus also contains a high-level language for specifying computations and writing functions. This document describes S-plus functions that integrate and extend existing S-plus functions for graphical display and analysis of time series data. These new functions allow the user to identify a model, estimate the model parameters, and forecast future values of a univariate time-series by using ARIMA (or Box-Jenkins) models. Use of these functions provides rapid feedback about time-series data and fitted models. These functions make it easy to fit, check and compare ARIMA models. Key Words: S-plus, ARIMA model, nonstationary time-series data, identification, estimation and diagnostic checking. 1 Introduction 1.1 Time-series data Time series data are collected and analyzed in a wide variety of applications and for many different reasons. One common application of time series analysis is for short-te...