Bayesian Ordination and Regression AnaLysis
Distributions available in boral
Correlation structure for latent variables
Including response-specific random intercepts in boral
Stochastic search variable selection (SSVS) in boral
Including species traits in boral
Bayesian Ordination and Regression AnaLysis (boral)
Fitting boral (Bayesian Ordination and Regression AnaLysis) models
Conditional log-likelihood for a fitted model
Log-likelihood for a model fitted with no latent variables
Marginal log-likelihood for a fitted model
Variance partitioning for a latent variable model
Caterpillar plots of the regression coefficients from a fitted model
lifecycle::badge("stable")Simulate a Multivariate response matrix
Dunn-Smyth Residuals for a fitted model
Extract Model Fitted Values for an boral object
Extract Deviance Information Criterion for a fitted model
Extract covariances and correlations due to shared environmental respo...
Highest posterior density intervals for a fitted model
Extract MCMC samples from models
Information Criteria for models
Additional Information Criteria for models
Extract residual correlations and precisions from models
Plot the latent variables from a fitted model
Write a text file containing a model for use into JAGS
Write a text file containing a model for use into JAGS
Plots of a fitted boral object
Predict using a model
Caterpillar plots of response-specific random effects from a fitted mo...
Summary of fitted boral object
Reformats output from a boral fit
Bayesian approaches for analyzing multivariate data in ecology. Estimation is performed using Markov Chain Monte Carlo (MCMC) methods via Three. JAGS types of models may be fitted: 1) With explanatory variables only, boral fits independent column Generalized Linear Models (GLMs) to each column of the response matrix; 2) With latent variables only, boral fits a purely latent variable model for model-based unconstrained ordination; 3) With explanatory and latent variables, boral fits correlated column GLMs with latent variables to account for any residual correlation between the columns of the response matrix.