Density-structured models & best practices for teaching in R
Description changed:
Best practices for teaching (and learning) R
The practice of teaching (and learning) R has changed dramatically in the past 10 years. In this session I will discuss how innovations such RStudio and tidyverse have created a more interactive and engaging experience for both learners and trainers. If you teach R, or are starting to learn R, come along and share your experiences on what approaches and resources you have found useful.
Speaker bio: Mark joined the University of Sheffield in October to start a Bioinformatics Core service having previously worked for Cancer Research Uk in Cambridge. He has been teaching R and data analysis for 10 years and recently qualified as a Software Carpentry instructor.
Using density-structured models to understand weed dynamics at large scales
A central problem in ecology is the mismatch between the scale at which we can easily collect data, and the scale at which we need to understand populations. This problem is caused by the expensive and time-consuming nature of data collection and amplified by the variable nature of natural systems across time and space. Using small-scale data with ‘traditional’ population models can therefore be problematic when trying to predict or describe large-scale population dynamics. Addressing this scale mismatch requires the expansion of data collection and the development of appropriate population models. Density-structured models provide rapid data collection via the summarisation of abundances by into discrete density ‘states’, and robust analysis through modelling dynamics as a function of transition probability between these categories. This talk will be a short tour of my PhD research, where I explored the use of density-structured models to investigate the dynamics of plant populations over large scales.
Speaker bio: Rob Goodsell is a postdoctoral researcher in APS working on large-scale ecological modelling. In this talk he will give us an overview of his PhD project [using rstan, tidyverse, ggplot2}.
