Run diagnostics for a state-space model fit
ppl_diagnose_ssm.RdThis function performs basic checks for an SSM fit, such as convergence of parameters and posterior predictive checks
Arguments
- fit
A JAGS state-space model fitted to CWAC data
- counts
A dataframe with CWAC counts ready for model fit. See
ppl_create_data_ssm- sp_code
SAFRING reference number of the species we want to analyse.
- config
A list with pipeline configuration parameters. See
configPipeline
Value
A list with Rhat values and posterior check statistics is returned. At them moment, we obtain Rhat values for all monitored parameters and three posterior check statistics: "Tmean" proportion of posterior simulations with mean greater than that observed in the data (we would like values close to 0.5), "Tsd" proportion of posterior simulations with sd greater than that observed in the data (we would like values close to 0.5), "Tdiff" mean difference between observed data and posterior simulations (we would like values close to 0).