ONLINE COURSE – Time Series Analysis and Forecasting using R and Rstudio (TSAF01) https://www.prstats.org/course/time-series-analysis-and-forecasting-using-r-... PR stats have a course on time series analysis and forecasting scheduled for October. Although this course does not use ecological datasets, all the skills learnt can be applied to ecological time series datasets, commonly encountered in long term studies of marine mammals. Instructor - Dr. Rafael De Andrade Moral 17th - 26th October Please feel free to share! In this six-day course (Approx. 35 hours), we provide a comprehensive practical and theoretical introduction to time series analysis and forecasting methods using R. Forecasting tools are useful in many areas, such as finance, meteorology, ecology, public policy, and health. We start by introducing the concepts of time series and stationarity, which will help us when studying ARIMA-type models. We will also cover autocorrelation functions and series decomposition methods. Then, we will introduce benchmark forecasting methods, namely the naïve (or random walk) method, mean, drift, and seasonal naïve methods. After that, we will present different exponential smoothing methods (simple, Holt’s linear method, and Holt-Winters seasonal method). We will then cover autoregressive integrated moving-average (or ARIMA) models, with and without seasonality. We will also cover Generalized Additive Models (GAMs) and how they can be used to incorporate seasonality effects in the analysis of time series data. Finally, we will cover Bayesian implementations of time series models and introduce extended models, such as ARCH, GARCH and stochastic volatility models, as well as Brownian motion and Ornstein-Uhlenbeck processes. Please email oliverhooker@prstatistics.com with any questions. -- Best wishes, Oliver Oliver Hooker PhD. PR stats