Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/gavinsimpson/waterloo2022
https://github.com/gavinsimpson/waterloo2022
Last synced: 23 days ago
JSON representation
- Host: GitHub
- URL: https://github.com/gavinsimpson/waterloo2022
- Owner: gavinsimpson
- Created: 2023-08-14T09:51:43.000Z (about 1 year ago)
- Default Branch: main
- Last Pushed: 2023-08-14T09:58:16.000Z (about 1 year ago)
- Last Synced: 2024-06-11T19:57:56.430Z (5 months ago)
- Language: HTML
- Homepage: https://gavinsimpson.github.io/waterloo2022/slides/
- Size: 49.1 MB
- Stars: 4
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Estimating trends in messy time series using generalized additive models
## Statistics & Biostatistic Seminar Series, Department of Statistics & Actuarial Science, University of Waterloo, 10th March
A defining characteristic of ecological and environmental time series is that they are messy. They're full of holes, rarely long enough but often too big, and they have a tendency to start and stop at inconvenient times and places. On top of this, the things these time series measure are inherently complex, the result of myriad interactions and processes operating throughout the biogeosphere.
In this talk I'll discuss how we use generalized additive models (GAMs) to estimate nonlinear spatio-temporal trends from these messy time series. I'll explain what modern GAMs are and how they learn from data using penalized splines, and illustrate how these models work with several spatio temporal examples drawn from my own research on environmental time series.