Explore data science applications in environment and ecology through this comprehensive lecture by Professor Ruth King from the University of Edinburgh. Delve into various data collection methods, including spatial, spatio-temporal, and individual-level data. Learn about hidden Markov models, state-space models, and Bayesian inference techniques. Examine real-world examples using ring-recovery and capture-recapture data, and understand the process of constructing statistical models. Discover the differences between classical and Bayesian approaches to parameter estimation and model choice. Gain insights into handling issues such as missing data and incorporating different forms of heterogeneity. Apply these concepts to practical examples in ecology and epidemiology, including analyses of capture-recapture and count data using Bayesian methods.
Overview
Syllabus
Introduction
Data collection
Spatial data
Spatio temporal data
Individual level data
Data analysis
Example 1: Ring-recovery data
Example 1: Assumptions
Example 1: Model parameters
Example 1: Statistical model
Example 2: Assumptions
Example 2: Statistical model
Decisions in constructing models
Discussion-building models for capture-recapture data
Discussion-building models for telemetry data
Classical approach
Bayesian approach
Bayesian parameter estimation
MCMC single update overview
Statistical analysis
Issue 1: Model choice
Issue 1: Classical model choice
Issue 1: Bayesian model choice
Example: Model choice
Statistical approaches
Example 1: Capture-recapture data - Bayesian analysis
Example 2: Count data - Bayesian analysis output
Taught by
Alan Turing Institute
