Online course: Time Series Analysis using regression techniques
This online course consists of 5 modules representing a total of approximately 40 hours of work. Each module consists of video files with short theory presentations, followed by exercises using real data sets, and video files discussing the solutions and R code. All video files are on-demand and can be watched online, as often as you want, at any time of the day, within a 12-month period.
You can ask course-related questions on the Discussion Board or in a live chatbox. The course fee includes a 1-hour face-to-face video chat with the instructors. During this meeting, you can ask any questions (e.g. about your own data analysis).
Some universities and institutes organize this course as a 5-day 'Live online teaching' course for 20-25 participants, typically from 09.00-16.00 (times may differ per university). Highland Statistics also runs it once or twice per year as an open course. In that case, the course contains 5 2-hour live web meetings in which we summarise some of the exercises.
You can do this course also with self-study.
The time series course starts with a short revision of data exploration and multiple linear regression. A non-technical introduction of generalised additive models (GAM) is provided. GAMs will be used to estimate long-term trends, seasonal patterns, covariate effects, and auto-regressive correlation. We also provide a short introduction to linear mixed-effects models and generalised linear mixed-effects models (GLMM) to analyse hierarchical data (e.g. short time series from the same core or site). GLMMs and generalised additive mixed-effects models (GAMM) are used to estimate trends, seasonality, covariate effects, and dependency in multivariate time series.
During the course, we will analyse time-series data sets containing continuous, binary, proportional, and count data. GLMs, GAMs, GLMMs, and GAMMs with the Gaussian, Poisson, negative binomial, Bernoulli, binomial, beta, gamma and Tweedie distributions are used. Throughout the course, we will use the mgcv and glmmTMB packages in R.
A detailed outline of the course is provided below. All exercises consist of a data set, a video describing the data and the questions, R solution code, and a video discussing the R solution file. Preparation material on data exploration and two exercises are provided.
- Revision exercise on multiple linear regression.
Short theory presentation on matrix notation.
Theory presentation 'Introduction to GAM'.
Three exercises to get familiar with GAM
- Theory presentation: How to include auto-regressive correlation in a regression model.
- Exercise showing how to fit a GLM with AR1 correlation in glmmTMB.
- Exercise on GAM with auto-regressive correlation applied to a regular spaced time-series data set.
- Exercise on GAM with auto-regressive correlation applied to an irregular spaced time-series data set.
- Exercise on detecting important changes in trends.
- Theory presentation on linear mixed-effects models.
- Exercise on linear mixed-effects models.
- Two exercises on the application of GAMM on time-series data sets.
- Theory presentation on distributions.
- Theory presentation: Revision of Poisson and negative binomial GLM.
- Revision exercise on Poisson GLM.
- Two exercises on the application of Poisson and negative binomial GAM applied to the univariate time series.
- Exercise on the application of Bernoulli GAM applied to a time series data set.
- Exercise on Bernoulli GAMM applied to a multivariate time-series data set.
- Exercise on beta GAMM applied to a time-series data set.
- Exercise on gamma GAM(M) applied to a time-series data set.
- Exercise on Tweedie GAM(M) applied to a time-series data set.
- Exercise on binomial GAM(M) applied to a time-series data set.
- Bonus 1: Theory presentation on data exploration.
- Bonus 2: Exercise on multiple linear regression analysis.
- Bonus 3: Exercise on GAMM applied to time series of tagged animals.
- Bonus 4: Exercise linear mixed-effects models using the bears and ants data set.
- Bonus 5: Exercise introducing the negative binomial (NB), generalised Poisson (GP) and Conway-Maxwell-Poisson GLMs for the analysis of puffin count data.
- Bonus 6: Exercise showing the application of Poisson and NB GLMM for the analysis of beehive time series.
- Bonus 7: What is DHARMa?
What data sets do we use?
During this course, you will see univariate time series (e.g. stable oxygen isotope data from cores, monthly temperature since 1600, the occurrence of whale strandings on North Sea coats since 1600, annual counts of Coots on Hawaii, the weekly number of puma visitations at a site, the proportion of seeds in florets of Trodia grass) and also multivariate time series (e.g. activity data from 18 tagged lynx, isotope data from 12 penguin colonies, temperature data from under the snow at 12 sites, survival of elephant calves from a large number of elephant mothers, weekly movement data of polar bears, bi-monthly algal Chlorophyll-a concentrations at 5 sites for 6 years, monthly coral coverage on 18 plates)
Some of these time series are regularly spaced and others are irregularly spaced in time. The examples cover continuous data, count data, absence/presence data, proportional data, and continuous data larger (or equal) to 0.
Free 1-hour face-to-face video meeting: The course fee includes a 1-hour face-to-face video meeting with one or both instructors. The meeting needs to take place within 12 months after the last live zoom meeting. You can discuss your own data but the statistical topics need to be within our field of expertise. The 1-hour needs to be consumed in one session and will take place at a mutually convenient time.
Discussion Board: You can use the Discussion Board to ask any questions related to the course material.
Pre-required knowledge: Working knowledge of R, data exploration and multiple linear regression. A revision of GLM and mixed-effects models are provided. This is a non-technical course.
Cancellation policy: What if you are not able to participate? Once participants are given access to course exercises with R solution codes, pdf files of certain book chapters, pdf files of PowerPoint or Prezi presentations and video solution files, all course fees are non-refundable and non-transferable to another participant.
Copyright: Sharing the access details of the course website or the pdf files of our course material is prohibited. Video files cannot be downloaded, but they can be watched in the same way as on Netflix.