- This event has passed.
Time series models for ecologists and climatologists (TSMC01)
10th May 2016 - 13th May 2016
This course will cover model-based time series analysis with a particular focus on applications in ecology and climatology. All methods will be illustrated using the free, open-source software package R. Time Series data are ubiquitous in the physical sciences, and models for their behaviour enable scientists to understand temporal dynamics and predict future values. Participants will be taught a wide range of suitable time series models for both discrete and continuous time systems. The course takes a foundational Bayesian approach, which will enable participants to have a deeper understanding of the models being fitted, and to estimate all unknown quantities with uncertainty. Participants are encouraged to bring their own data sets for discussion with the course tutors.
Research postgraduates, practicing academics in climatology, meteorology, conservation and environmental management, and environmental professionals in government and industry.
A mixture of lectures and hands-on practical’s. Introductory lectures on the concepts and refreshers on R usage. Intermediate-level lectures interspersed with hands-on mini practicals and longer projects. Finally, round-table discussions about the analysis requirements of attendees. Data sets for computer practicals will be provided by the instructors, but participants are welcome to bring their own data.
Assumed quantitative knowledge
A basic understanding of statistical concepts. Such as regression modelling and generalised linear models. Some understanding of Bayesian Statistics is recommended but will be covered during the introductory sessions.
Assumed computer background
Previous experience with data analysis using R is required. Ability to import/export data, manipulate data frames, fit basic statistical models & generate simple exploratory and diagnostic plots.
Equipment and software requirements
A laptop/personal computer with a working version or R, RStudio and JAGS installed. R, RStudio and JAGS are supported by both PC and MAC and can be downloaded for free by following these links.
Further information on interfacing JAGS and R using the runjags package can be found from the following website: http://runjags.sourceforge.net/.
It is essential that you come with all necessary software and packages already installed (you will be sent a list of packages prior to the course) internet access may not always be available.
UNSURE ABOUT SUITABLILITY THEN PLEASE ASK email@example.com
Monday 9thMeet at the Tullie Inn, Balloch at approximately 18:30 before being taken by minibus to SCENE (Download directions PDF).Tuesday 10th – Classes from 09:00 to 17:00
Class 1: Introduction; some example time series datasets; prediction vs explanation.
Class 2: An introduction to Bayesian Statistics.
Class 3: The AR(1) model.
Practical: revision on using R to load data, create plots and fit statistical models.
Round table discussion: understanding the output from a Bayesian model.
Wednesday 11th – Classes from 09:00 to 17:00
Class 1: ARMA models for real data.
Class 2: ARIMA and sARIMA modelling.
Practical: An introduction to the Bayesian modelling language JAGS.
Round table discussion: understanding and running a JAGS model.
Thursday 12th – Classes from 09:00 to 17:00
Continuous Time Series Modelling.
Class 1: Brownian Motion and its application to real data sets.
Class 2: An introduction to Stochastic Volatility Modelling.
Practical: Fitting continuous time models in JAGS.
Round table discussion: Issues of continuous vs discrete time.
Friday 13th – Classes from 09:00 to 16:00
Advanced Times Series Models.
Class 1: Multivariate models.
Class 2: Fractional differencing and models using differential equations.
Practical: Running advanced models in JAGS.
Round table discussion: Bring your own data set.