5 very useful general stats course being offered by PR statistics.

Please feel free to share.


Bayesian Data Analysis (BADA02) 
https://www.prstatistics.com/course/bayesian-data-analysis-bada02/
ABOUT THIS COURSE
Bayesian methods are now increasingly widely in data analysis across most scientific research fields. Given that Bayesian methods differ conceptually and theoretically from their classical statistical counterparts that are traditionally taught in statistics courses, many researchers do not have opportunities to learn the fundamentals of Bayesian methods, which makes using Bayesian data analysis in practice more challenging. The aim of this course is to provide a solid introduction to Bayesian methods, both theoretically and practically. We will begin by teaching the fundamental concepts of Bayesian inference and Bayesian modelling, including how Bayesian methods differ from their classical statistics counterparts, and show how to do Bayesian data analysis in practice in R. We then provide a solid introduction to Bayesian approaches to these topics using R and the brms package. We begin by covering Bayesian approaches to linear regression. We will then proceed to Bayesian approaches to generalized linear models, including binary logistic regression, ordinal logistic regression, Poisson regression, zero-inflated models, etc. Finally, we will cover Bayesian approaches to multilevel and mixed effects models. Throughout this course, we will be using, via the brms package, Stan based Markov Chain Monte Carlo (MCMC) methods.


Reproducible and Collaborative Data Analysis with R (RACR01)
https://www.prstatistics.com/course/reproducible-and-collaborative-data-analysis-with-r-racr01/
ABOUT THIS COURSE
The computational part of a research is considered reproducible when other scientists (including ourselves in the future) can obtain identical results using the same code, data, workflow and software. Research results are often based on complex statistical analyses which make use of various software. In this context, it becomes rather difficult to guarantee the reproducibility of the research, which is increasingly considered a requirement to assess the validity of scientific claims. Moreover, reproducibility is not only important for findings published in academic journals. It also becomes relevant for sharing analyses within a team, with external collaborators and with one’s supervisor. During this three-day course, the participants will be introduced to a suite of tools they can use in combination with R to make reproducible the computational part of their own research. A strong emphasis is given to collaboration, and participants will learn how to set up a project to work with other people in an efficient way.


Nonlinear Regression using Generalized Additive Models (GNAM02)
https://www.prstatistics.com/course/nonlinear-regression-using-generalized-additive-models-gamr02/
ABOUT THIS COURSE
This course will cover introductory hierarchical modelling for real-world data sets from a Bayesian perspective. These methods lie at the forefront of statistics research and are a vital tool in the scientist’s toolbox. The course focuses on introducing concepts and demonstrating good practice in hierarchical models. All methods are demonstrated with data sets which participants can run themselves. Participants will be taught how to fit hierarchical models using the Bayesian modelling software Jags and Stan through the R software interface. The course covers the full gamut from simple regression models through to full generalised multivariate hierarchical structures. A Bayesian approach is taken throughout, meaning that participants can include all available information in their models and estimates all unknown quantities with uncertainty. Participants are encouraged to bring their own data sets for discussion with the course tutors.


Time Series Data Analysis (TSDA02)
https://www.prstatistics.com/course/online-course-time-series-data-analysis-tsda02/
ABOUT THIS COURSE
This course covers introductory modelling for the analysis of time series data. The main focus of the course is on data observed at regular (discrete) time points but later modules cover continuously-observed data. The methods are presented both at a theoretical level and also with practical examples where all code is available. The practical classes include instructions on how to use the popular forecast package. The second half of the course looks at Bayesian time series analysis which is extremely customisable to bespoke data analysis situations.


Introduction to generalised linear models using R and Rstudio (IGLM05)
https://www.prstatistics.com/course/introduction-to-generalised-linear-models-using-r-and-rstudio-iglm05/
ABOUT THIS COURSE
In this two day course, we provide a comprehensive practical and theoretical introduction to generalized linear models using R. Generalized linear models are generalizations of linear regression models for situations where the outcome variable is, for example, a binary, or ordinal, or count variable, etc. The specific models we cover include binary, binomial, ordinal, and categorical logistic regression, Poisson and negative binomial regression for count variables. We will also cover zero-inflated Poisson and negative binomial regression models. On the first day, we begin by providing a brief overview of the normal general linear model. Understanding this model is vital for the proper understanding of how it is generalized in generalized linear models. Next, we introduce the widely used binary logistic regression model, which is is a regression model for when the outcome variable is binary. Next, we cover the ordinal logistic regression model, specifically the cumulative logit ordinal regression model, which is used for the ordinal outcome data. We then cover the case of the categorical, also known as the multinomial, logistic regression, which is for modelling outcomes variables that are polychotomous, i.e., have more than two categorically distinct values. On the second day, we begin by covering Poisson regression, which is widely used for modelling outcome variables that are counts (i.e the number of times something has happened). We then cover the binomial logistic and negative binomial models, which are used for similar types of problems as those for which Poisson models are used, but make different or less restrictive assumptions. Finally, we will cover zero inflated Poisson and negative binomial models, which are for count data with excessive numbers of zero observations.

########################################################################

To unsubscribe from the ECOL-AGRIC list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=ECOL-AGRIC&A=1

This message was issued to members of www.jiscmail.ac.uk/ECOL-AGRIC, a mailing list hosted by www.jiscmail.ac.uk, terms & conditions are available at https://www.jiscmail.ac.uk/policyandsecurity/