This book will show you how to use R to model and forecast catch time series using a variety of standard forecasting models.

  • Time-varying regression
  • Box-Jenkins (ARIMA) models
  • Exponential smoothing
  • Multivaritate regression with ARMA errors
  • ARMA models with covariates (ARMAX)
  • Seasonal ARIMA models
  • Seasonal exponential smoothing models

In addition to model fitting, model diagnostics, forecast diagnostics and accuracy metrics will be covered along with uncertainty metrics.

The focus of this book is on analysis of univariate time series. However multivariate regression with autocorrelated errors and multivariate autoregressive models (MAR) will be covered more briefly. For an indepth discussion of multivariate autoregressive models and multivariate autoregressive state-space models, see Holmes, Ward and Scheuerell (2018).

As a work of the United States government, this project is in the public domain within the United States of America. Additionally, we waive copyright and related rights in the work worldwide through the Unlicense public domain dedication. Unlicense public domain dedication.