This advanced workshop aims at introducing students to linear modelling methods that are commonly used in Ecology today. The course is divided into a series of modules that build on each other towards more complex linear models, starting from simple linear models and generalised linear models, to mixed effects models with grouped random effects, and linear models with generalised least squares for random effects that cannot be grouped. The underlying linear models are explained so experience with linear algebra will be very helpful though not essential to complete the course

01

Dates

22 – 28 June, 2024；Check-in on 22 June

02

Venue

XTBG, Menglun, Mengla, Xishuangbanna, Yunnan, 666303

03

Organiser

Xishuangbanna Tropical Botanical Garden, Chinese Academy of Sciences (XTBG)

04

Registration

Registration is open until 15:00, 15 June, 2024

05

Target students

Graduate students, young scholars, practitioners etc. in related fields such as Ecology. No more than 30 person.

06

Fee

(1) 1800 RMB/person (including lecture room, materials etc.

(2) Transportation between your organization and XTBG, hotel and food during the workshop etc. are self-care.

07

Course credits

This workshop has obtained accreditation from the University of Chinese Academy of Sciences (UCAS). Graduate students from UCAS and institutions affiliated to Chinese Academy of Sciences will receive 2 course credits towards their degree program upon completion of the study. Other participants may be able to arrange recognition of these credits by their home institutions.

08

Workshop outline

Day 1

Check-in 

Day 2

Module 1: Classical methods 

The first day revises basic statistical concepts and classical methods, and links them via the linear model and the generalised linear model.

Lecture 1: Linear models  

Lecture 2: General linear models

Day 3~5

Module 2: Modelling with grouped non-independent data: mixed effects models 

The bulk of this course deals with the problem of non-independence of data and how to fix it appropriately using linear models. The main issue is recognizing how individual data points may be related to one another, which means that they share information. Our statistical tests assume that there is no relatedness between residuals generated by fitting models to the data.  In this first section we deal with situations where that relatedness can be reasonably described by grouping related data points. This has the dual benefits of improving residual independence and also accounting for additional information in the data, potentially improving our test sensitivity. In doing this, we also recognise that accounting for this relatedness may not be essential to our questions; so we treat these additional model terms as random effects, and the predictors which are core to our questions we treat as fixed effects. Models with both fixed and random effects are called mixed effects models.

Lecture 3: Intro to linear mixed models

Lecture 4: LMM types

Lecture 5: Inference with LMMs

Lecture 6:  Predictions with LMMs

Lecture 7: GLMMs

Day 6

Module 3: Modelling with non-independent data that cannot be grouped: generalized least squares 

Certain types of data are non-independent, but the non-independence cannot be corrected by grouping, e.g. autocorrelation in spatial and temporal data, which depend on individual pairwise distances between points. In this module we introduce methods for dealing non-independence between data points using correlation matrices.

Lecture 8: Generalised least squares

Lecture 9: Phylogenetic regression

Day 7

Lecture 10: The Animal model

Recognising that all data could contain both residual relatedness due to groups and due to pairwise relatedness means we need a model that can handle both situations. The Animal model was devised to deal with these situations.

Module 4: Intro to Bayesian models

Lecture 11: Bayesian Basics

09

Requirements for applicants

As an advanced course, it has some certain pre-requisites for admission.

(1) Participants should have seen at least one semester of statistics at university or college. Students who have completed the AFEC stats module will also be considered. 

(2) Participants should be familiar with R. Students familiar with S and SAS may also apply. Please provide evidence in line with each of these requirements when applying for the course.

(3) The course is divided into a series of modules that build on each other towards more complex linear models, starting from simple linear models and generalised linear models, to mixed effects models with grouped random effects, and linear models with generalised least squares for random effects that cannot be grouped. The underlying linear models are explained so experience with linear algebra will be very helpful though not essential to complete the course.

10

Instructors

Prof. Kyle Tomlinson

PI of Community Ecology and Conservation Group of XTBG, works on landscape conservation, forest ecology, savanna ecology, and functional trait diversity. Kyle is an experienced statistics instructor and has been invited to run statistic workshops during ATBC-Asia chapter annual meetings, and during Advanced Fieldcourse in Ecology and Conservaton (XTBG's annual international training program) since 2013. Since 2020 he has been the associate editor of the Journal of Ecology.

Prof. Gbadamassi G. O. Dossa

He is from Forest Canopy Ecology Group of XTBG, working on biodiversity, forest ecosystem functioning, plant functional traits effect on carbon and nutrient cycling. Dossa designed and taught several intensive, high-level courses or workshops for students including Open / Reproducible science, Statistics (beginner level), R language for statistics, Critical scientific reading, Nutrient cycling etc. Since 2017, he has been one of the main instructors or teaching assistants of Statistics (advanced level), helped to engage students in advanced statistics with different teaching methods and feedback to make it more enjoyable.

Dr. FAN Huan

Graduated with a Ph.D. from the University of Wisconsin-Madison, completed her postdoctoral training with Professor Kyle Tomlinson at XTBG and is currently a research scientist in the plant breeding in Singapore. Her research mainly focuses on genome-wide association studies. She has extensive experience in data analysis and teaching, and has been teaching the Bayesian section for this training class since 2020.