Motivation | Workflow in DHARMa | Installing, loading and citing the package | Calculating scaled residuals | Plotting the scaled residuals | Goodness-of-fit tests on the scaled residuals | Simulation options | Refit | Conditional vs. unconditional simulations | Integer treatment / randomization | Calculating residuals for groups or subsets | Reproducibility notes, random seed and random state | Interpreting residuals and recognizing misspecification problems | General remarks on interpreting residual patterns and tests | Recognizing over/underdispersion | Residual patterns of over/underdispersion | Formal tests for over/underdispersion | Zero-inflation / k-inflation or deficits | Residual patterns | Formal tests for zero-inflation | Testing generic summary statistics, e.g. for k-inflation or deficits | Heteroscedasticity | Detecting missing predictors or wrong functional assumptions | Residual correlation structures (temporal, spatial, phylogenetic) | Case studies and examples | Budworm example (count-proportion n/k binomial) | Owl example (count data) | Notes on particular data types | Poisson data | Proportional data | Binomial data | Supported packages and frameworks | lm and glm | lme4 | mgcv | gamm4 | glmmTMB | spaMM | GLMMadaptive | phylolm | phyr | brms | Unsupported packages | Importing external simulations (e.g. from Bayesian software or unsupported packages)

Basic workflow | Example in Jags | Exercise | Conditional vs. unconditional simulations in hierarchical models | Statistical differences between Bayesian vs. MLE quantile residuals

Interfacing a model with BT - step-by-step guide | Step 1: Create a runModel(par) function | Case 1 - model programmed in R | Case 2 - compiled dll, parameters are set via dll interface | Case 3 - model programmed in C / C++, interfaced with RCPP | Case 4 - compiled executable, parameters set via command line (std I/O) | Case 5 - compiled model, parameters set via parameter file or in any other method | Case 6 - compiled model, parameters cannot be changed | Step 2: Read back data | Step 3 (optional) - creating an R package from your code | Speed optimization and parallelization | Easy things | Difficult things | Parallelization | Within sampler parallelization | Running several MCMCs in parallel | Thread safety

Quick start | Install, load and cite the package | The Bayesian Setup | Run MCMC and SMC functions | Convergence checks for MCMCs | Summarize the outputs | BayesianSetup Reference | Reference on creating likelihoods | Parallelization of the likelihood evaluations | 1. In-build parallelization: | 2. External parallelization | 3. Multi-core and cluster calculations | Reference on creating priors | Creating priors | No prior set | Set min/max values | Pre-defined priors | Creating user-defined priors | Creating a prior from a previous MCMC sample | MCMC sampler reference | The different MCMC samplers | The Metropolis MCMC class | Standard MH MCMC | Standard MH MCMC, prior optimization | Adaptive MCMC, prior optimization | Standard MCMC, prior optimization, delayed rejection | Adaptive MCMC, prior optimization, delayed rejection | Standard MCMC, prior optimization, Gibbs updating | Standard MCMC, prior optimization, gibbs updating, tempering | Differential Evolution MCMC | DREAM sampler | T-walk | Non-MCMC sampling algorithms | Rejection samling | Sequential Monte Carlo (SMC) | Bayesian model comparison and averaging | Example | Model comparison via Bayes factor | Model Comparison via WAIC