1. The problem: one line, two stories | 2. Why quantiles, and why a mixture | 3. An illustrative dataset | 4. Fitting your first model | 5. Visualizing the estimates | 5.1 The two regimes | 5.2 A coefficient comparison | 6. Who is in which group? | 7. Diagnostics: can you trust it? | 7.1 The component error densities | 7.2 Two engines agree | 7.3 A caveat worth knowing | 8. Inference done right | 9. Beyond the median: the whole distribution | 10. How many groups? | 11. Reporting and reproducibility | Citation | References

A two-regime example | A first picture | Where to next | References

1. The question | 2. The model in one page | 3. An illustrative dataset | 4. Fitting your first model | 5. Interpreting the gate | 6. Getting the uncertainty right | 7. Does membership shift across the distribution? | 8. Reading the fit in pictures | 9. Diagnostics: can you trust it? | 10. Practical guidance and pitfalls | 11. How mixqrgate relates to other tools | 12. Reporting and reproducibility | References

A concomitant gate | Does the gate vary with the quantile? | Notes | References

1. The problem: a treatment people choose | Why covariates enter | 2. The four core estimators in one picture | Why the paper (and the package default) prefers IPWRA | 3. A worked example | The naive answers fail | The drlate answer | Seeing double robustness work | 4. Checking the design: diagnostics | Overlap | Covariate balance | Weight distributions | A formal balance test | Profiling the compliers | 5. Choosing models and options | Outcome and treatment families | Instrument propensity score flavors | LATT: the effect for treated compliers | 6. Abadie's kappa: the weighting-estimator menu | 7. When the instrument is weak: Fieller confidence sets | 8. Bootstrap inference | 9. How much does the estimator choice matter? | 10. Do you even need the instrument? The DR Hausman test | 11. Coming from Stata | References

Overview | Joint inference | Example | Other estimators | Abadie-kappa weighting estimators | LATT, other model families, and IPT | Clustered standard errors and weights | Diagnostics | Inference beyond the default sandwich | The DR Hausman test of unconfoundedness | Comparing estimators | Replicating the Stata examples | Citation | References

Introduction | The R Synthetic Data Ecosystem | Where rsdv fits | The Gaussian Copula: A Practitioner's Introduction | Getting Started | Installation | A five-line synthesis | Describing Your Data: The Metadata System | Fitting and Sampling | Choosing marginal distributions | Conditional Sampling | Evaluating Quality | Column-level similarity | Correlation structure | Marginal distributions | Diagnostic checks | Evaluating Privacy | Adding Constraints | Handling Missing Data | What rsdv does by default | When to pre-impute | Considerations and Caveats | Citation | References

1. Why model quantiles? | Why multilevel? | Why Bayesian? | 2. The model in one page | 3. Your first model | 4. Many quantiles at once | 5. Priors and the scale parameter | 6. Getting the uncertainty right | 7. Correlated random effects | 8. Diagnostics: is the fit trustworthy? | 9. Visualising results | 10. Practical guidance and pitfalls | 11. How bqmm relates to other tools | References

A first model | Several quantiles at once | Valid inference

Random intercepts | Random slopes | Nested and crossed grouping | Practical notes

The problem | The correction | Why not the plain Koenker sandwich? | Scope and caveats | References

Why switch? | Side-by-side comparison | synthpop workflow | rsdv workflow | Key differences