Preliminaries | Exam data | Residual analysis | Influence analysis | Diagnostics for variance components

Installation | Details | Q-Q plot | P-P plot | Usage | Shiny App | References

Overview | hlm_influence() function | Introduction | Obtain influence diagnostics at different levels with level | Investigate deletion of specific observations or groups with delete | Select different types of leverage with leverage argument | Choose between approximations or full refits with approx | na.action and the data argument | hlm_augment function | lme objects from nlme package | Important differences for lme objects | Cook's distance values comparison to other packages | The car package | The lme4 package | lme4 approximation | lme4 full refit

Overview | The Data | The Call | 1. The Parametric Bootstrap | Examples | bootstrap() output | 2. The Cases Bootstrap | 3. The Residual Bootstrap | 4. The Wild Bootstrap | 5. The Random Effects Block (REB) Bootstrap | 6. Output Options | The summary() function | Example | The print() Function | The confint() Function | The plot() function | 7. Bootstrapping with GLMMs | 8. Parallelization (User Extension) | The Idea | Performance | Implementation | 9. Future Directions (Developer Extensions) | References

HLMdiag: a diagnostic tool for hierarchical (multilevel) linear models | hlm_resid() function | Types of residuals in multilevel models | Example Data | hlm_resid() usage | General usage | standardize and include.ls | Higher-level residuals | na.action and data | Three-level models | Using hlm_resid() in context | pull_resid() | hlm_augment() | References