Chapter 15
Time Series Analysis
Fit Time Series Models and Transfer Functions
The Time Series platform lets you explore, analyze, and forecast univariate time series. A time series is a set y1, y2, ... ,yN of observations taken over a series of equally-spaced time periods. The analysis begins with a plot of the points in the time series. In addition, the platform displays graphs of the autocorrelations and partial autocorrelations of the series. These indicate how and to what degree each point in the series is correlated with earlier values in the series.
You can interactively add:
Variograms
a characterization of process disturbances
AR coefficients
autoregressive coefficients
Spectral Density Plots
versus period and frequency, with white noise tests.
These graphs can be used to identify the type of model appropriate for describing and predicting (forecasting) the evolution of the time series. The model types include:
ARIMA
autoregressive integrated moving-average, often called Box-Jenkins models
Seasonal ARIMA
ARIMA models with a seasonal component
Smoothing Models
several forms of exponential smoothing and Winter’s method
Transfer Function Models
for modeling with input series.