Generalized additive modeling

Knowledge graph centered on Generalized additive modeling with 76 nodes and 119 connections. Top connected: Rocky Mountain Biological Laboratory, Gothic, Plant functional trait change across a warming tun, Two-form Bumble Bee, Bumble Bees.

Method synopsis

Hierarchical Bayesian models predicting vital rates from individual and neighborhood characteristics, with population-level random effects and model selection via WAIC.

Synthesized from method descriptions across 8 papers using this protocol.

Procedure from a recent peer-reviewed implementation

Steps below were extracted from the most recent peer-reviewed implementation of this protocol in the corpus — A multi-year case study highlighting the influence of hydrological conditions on epidemic dynamics in a natural plant pathosystem (2025), Oikos. The protocol was originally introduced by Assessing the forecastability & forecast skill of models to predict sex ratios of Valeriana edulis (2020). Implementations in other papers (listed below) may differ.

1. Apply generalized additive models
   GAMs were fit using the gam() function from the 'mgcv' package with family='gaussian' and method='REML' to explore empirical relationships between epidemiological parameters and measures of temperature and humidity, using mean daily values averaged from 1 July to 25 July and annual snow cover duration in eleven epidemics from BT, CC, GM sites
   Quantities: Eleven epidemics from three sitesDuration: Statistical analysis phaseConditions: Computational analysis
   Equipment: R mgcv package, GAM statistical software

Typical Equipment

• R statistical software v.4.4.1
• rstanarm package v.2.32.1
• Stan v.2.32.6

Output Measurements

• vital rate predictions
• model fit statistics
• parameter estimates

Papers Using This Protocol (8)

Species Studied With This Protocol

Places Where Used