Chapter 1 Introduction

There are many approaches for forecasting from time series alone–meaning without any covariates or exogenous variables. Examples are the approaches used in the following papers.

Stergiou and Christou 1996

  • Time-varying regression
  • Box-Jenkins models, aka ARIMA models
  • Multivariate time-series approaches
    • Harmonic regression
    • Dynamic regression
    • Vector autoregression (MAR)
  • Exponential smoothing (2 variants)
  • Exponential surplus yield model (FOX)

Georgakarakos et al. 2006

  • Box-Jenkins models, aka ARIMA models
  • Artificial neural networks (ANNs)
  • Bayesian dynamic models

Lawer 2016

  • Box-Jenkins models, aka ARIMA models
  • Artificial neural networks (ANNs)
  • Exponential Smoothing (6 variants)

This course will focus on three of these methods: time-varying regression, ARIMA models and Exponential smoothing models. These will be shown with and without seasonality. Methods which use covariates, or exogenous variables, will also be addressed.