BayesSummer

Bayes Summer School - August, 14th to 18th 2017

Projects

Participants Title Notes
Annie-Claude Belisle Modeling cumulative impacts of climate change and natural resource extraction on cultural landscapes. Participatory modeling is assed with two indigenous communities from boreal Quebec. Local experts will provide conceptual models of landscape values for seven ecosystem services and prior distributions for parameters.
Francois Rousseu Taking spatial correlation into account when measuring the effect of landscape on the occurence of attacks on cattle by Jaguars.  
Chunchao Zhu Disentangling the effects of plant traits, neighborhoods and topography on herbivore damage  
Alan Cohen Modeling longitudinal change in physiology during aging Using aggregate physiological indices of the aging process in human cohort data (2-4 visits per individual over 10 years), we will model individual aging rates using an exponential model of change.
Mathilde Besson, Madeleine Proulx, Azénor Bideault Réseau tri-partite le long d’un gradient latitudinal  
Amaël Le Squin Estimation of species-specific mortality parameters for 14 North-Eastern american trees Individual-based forest dynamics models require parameters (such as growth, mortality, fecundity…) to run simulations. Here, I want to parametrise mortality. Priors from Purves et. al. (2008) are available for some species thence Bayesian statistics seem well designed for this study.
Xavier Guilbault-Mayers Analyse de l’interaction entre les traits fonctionnels végétales et la fertilité des sols. Document complémentaire
Alexis Carteron Soil constraints on sugar maple migration from temperate to boreal forest Abiotic and biotic effects on survival and growth of seedlings assessed using a zero-inflated model
Louis Donelle Accounting for intraspecific trait variation in community trait-environment relationship: A Bayesian approach. Document complémentaire
Elliot Dreujou Create an anthropogenic stress map on benthic communities Starting from a dataset of human usage of the environment (gathered by expert consultation and quantitative sampling), I wish to model the spatial distribution of several anthropogenic stressors that affect benthic communities.
Peter Morison Analysis of conservation status of species in Canada The Canadian Wildlife Service recently released a report on the conservation status of 29 848 species in Canada among 34 different taxonomic groups. This project will seek to explain the level of risk of these species based on past assessments of risk, jurisdiction, taxonomic group, and (possibly) phylogenetic relationships.
Lucas Deschamps, Florence Tauc Is diversity of chlorophyll distribution related to light variability in peatland communities? We have observed a high, non-random, diversity in distribution of leaf chlorophyll concentration of wetlands communities. We have made the hypothesis that this variability was due to the fine-scale adaptation of leaves, depending unpon the geometry of the canopy. We will approach this hypothesis by linking the variance of light attenuation and the diversity metrics calculated for each plot. This imply to use multilevel modelling to account for the daily variation of light captured by 7 light recorder placed in each plot.

Material

  Lesson Speaker
Day 1 Introduction Dominique Gravel
Day 1 Probability theory Kevin Cazelles
Day 2 Likelihood Dominique Gravel
Day 3 MCMC Guillaume Blanchet
Day 4 Hierarchical modelling Guillaume Blanchet
Day 4 Solution to Exercices Guillaume Blanchet
Day 4 HMSC presentation Guillaume Blanchet
Day 5 Model comparison Matthew Talluto
Day 5 Hooten & Hobbs (2015) on model selection Matthew Talluto

Pick up your distribution: McLaughlin - 2014 - A compendium of common probability distributions

How to use a natural conjugate distribution - an example

Back to Github

Books

  1. Gelman, A. et al. Bayesian Data Analysis, Third Edition. (CRC Press, 2013)

  2. Gelman, A. & Hill, J. Data Analysis Using Regression and Multilevel/Hierarchical Models. (Cambridge University Press, 2006)


Informations générales pour l’école d’été en statistiques Bayésienne pour écologistes

English follows French

Transport et Hébergement

L’école d’été se déroulera à Orford Musique au 3165, chemin du Parc Orford (Google Maps) du 14 au 18 août 2017.

Si vous désirez vous rendre sur le lieu par autobus, la compagnie Limocar (https://limocar.ca/) offre des services à destination de Magog. Si vous utilisez ce mode de transport, assurer vous donc de nous contacter (ecole.biologie@usherbrooke.ca) à l’avance avec votre horaire et lieu d’arriver pour être certain qu’on puisse vous rencontrer à la bonne place et au bon moment.

Lors de votre séjour à Orford Musique, la literie ainsi que des serviettes de bains vous seront fournies. Il faudra par contre apporter votre gel douche et votre shampoing.

Des détails supplémentaires à propos d’Orford Musique sont disponibles ici.

Un plan du site d’Orford Musique se trouve ici.

Arrivé sur le lieu

Vous pouvez arriver sur le site le dimanche 13 août. S’il-vous-plaît, communiquer avec nous (ecole.biologie@usherbrooke.ca) si vous désirez bénéficier de la nuit du 13 août. Il n’y aura aucun frais mais pour des questions de logistique nous souhaiterions en être avisé.

La programmation de la semaine se trouve ici (seulement en anglais).

Matériels de cours à apporter

Lecture

Avant le début de l’école d’été, les étudiants doivent avoir lue les cinq articles suivant:

Choy, S. L., R. O’Leary, et K. Mengersen. 2009. Elicitation by design in ecology: using expert opinion to inform priors for Bayesian statistical models. Ecology 90:265–277. (PDF)

Cressie, N., C. A. Calder, J. S. Clark, J. M. V. Hoef, et C. K. Wikle. 2009. Accounting for uncertainty in ecological analysis: the strengths and limitations of hierarchical statistical modeling. Ecological Applications 19:553–570. (PDF)

Dorazio, R. M. 2016. Bayesian data analysis in population ecology: motivations, methods, and benefits. Population Ecology 58:31–44. (PDF)

Ellison, A. M. 1996. An introduction to Bayesian inference for ecological research and environmental decision-making. Ecological Applications 6:1036–1046. (PDF)

Ludwig, D. 1996. Uncertainty and the assessment of extinction probabilities. Ecological Applications 6:1067–1076. (PDF)

Activités extracurriculaires

Le site d’Orford Musique est exceptionnel. Il est placé directement au pied du Parc National du Mont Orford. Il vous sera donc possible de faire de la randonnée et de la course à pied sur ces sentiers à l’extérieur des périodes de cours.

Nous aurons aussi la possibilité de prendre un verre (avec ou sans alcool) autour d’un feu de camp en soirée. Même si certains breuvages alcoolisés peuvent être acheté sur site, vous pouvez aussi en apporter. Si vous avez une guitare et que vous êtes à la recherche d’un public, nous vous invitons à l’apporter !


General informations concerning the Bayesian statistics for ecologists summer school

Transportation and housing

The summer school will be held at Orford Musique located at 3165, chemin du Parc Orford (Google Maps), from August 14 to 18 2017.

If you need to get to Orford Musique by bus, Limocar (https://limocar.ca/) offers rides to Magog. If you are using this mean of transportation, make sure to contact us (ecole.biologie@usherbrooke.ca) in advance with your scheduled arrival to make sure he can meet you at the right place and time.

During your time at Orford Musique, beddings and towels will be provided. However, you will need to bring shower soap and shampoo.

Additional details about Orford Musique are available here (only in French).

A map of the site of Orford Musique is available here.

Arrival on site

You can arrive on site Sunday August 13. Please contact us (ecole.biologie@usherbrooke.ca) if you plan on passing the night of August 13 at Orford Musique. There will not be any additional fees related to this extra night, however, for logisitic reasons we prefer to be aware of who will arrive when.

Details of the weekly schedule can be found here

Class material to bring

Reading material

Before the beginning of the summer school, students are required to read the following five papers:

Choy, S. L., R. O’Leary, and K. Mengersen. 2009. Elicitation by design in ecology: using expert opinion to inform priors for Bayesian statistical models. Ecology 90:265–277. (PDF)

Cressie, N., C. A. Calder, J. S. Clark, J. M. V. Hoef, and C. K. Wikle. 2009. Accounting for uncertainty in ecological analysis: the strengths and limitations of hierarchical statistical modeling. Ecological Applications 19:553–570. (PDF)

Dorazio, R. M. 2016. Bayesian data analysis in population ecology: motivations, methods, and benefits. Population Ecology 58:31–44. (PDF)

Ellison, A. M. 1996. An introduction to Bayesian inference for ecological research and environmental decision-making. Ecological Applications 6:1036–1046. (PDF)

Ludwig, D. 1996. Uncertainty and the assessment of extinction probabilities. Ecological Applications 6:1067–1076. (PDF)

Extracurricular activities

Orford Musique is an exceptional site located at the base of Mont Orford National Park. As such, it will be possible for you to hike in the numerous path outside of class hours.

It will also be possible to have a drink (alcoholic or not) around a campfire in the evening. Even though some alcoholic beverages can be bought on site, you can also bring your own. Also, feel free to bring your guitar!