Introduction | Data format | Start with estimate_population() | Specifying log-linear models | BIC model selection | Which models are considered? | Bootstrap inference after BIC selection | Restricting model selection within the bootstrap | Sparse capture data | The stepwise method | Bootstrap inference after stepwise selection | Fitting a specified model | The Bayesian thresholding method | Which method should I use? | Looking inside the result | References

Introduction | The Bayesian-threshold procedure | Bayesian thresholding | MCMC estimation | MCMC settings and posterior output | A basic analysis | Improper priors | Four-list interactions | Checking existence with improper interaction priors | Examples with improper priors | Summary | References

Introduction | BIC model selection | Inspecting the fit | Choosing the interaction order | Bootstrap inference | Practical bootstrap inference | Reading the confidence limits | Choosing the retained model set | Summary | References

Capture histories | Encoding capture histories | Parents, children, ancestors and descendants | Hierarchical log-linear models | Describing a hierarchical model | Neighbouring hierarchies | Generating hierarchical models | Sparse and masked interactions | Implications for the log-linear model | Appendix: The number of hierarchical models | Reference

Introduction | Running the reproduction code | Applications and published results | Table 2: Kosovo | Table 3: Korea | Figure 1: Five-list UK data | Figure 2: Five-list New Orleans data | Further investigations from the 2024 paper | Table 4: Location of the bootstrap BIC minimum | Table 5: Degree-2 ranking for Kosovo | Figure 3: Degree-1 and degree-2 rankings | Figure 4: Rank of the bootstrap-optimal model | Table 6: Downhill model search | Summary | References

Introduction | Sparse capture data and extended maximum likelihood | Existence and identifiability | The stepwise model-selection procedure | Allowing higher-order interactions | Bootstrap inference | Applications and published results | New Orleans: Section 4.1 | Five-list New Orleans data | Western site data: Section 4.2 | Simulation studies | Comparison with the BIC approach: Section 3.4 | Choosing the threshold: Section 4.3 | Relation to the current package | References

Capture histories | Encoding capture histories | Parents, children, ancestors and descendants | Hierarchical log-linear models | Describing a hierarchical model | Neighbouring hierarchies | Generating hierarchical models | Sparse and masked interactions | Implications for the log-linear model | Appendix: The number of hierarchical models | Reference

Introduction | Data format | Start with estimate_population() | Specifying log-linear models | BIC model selection | Which models are considered? | Bootstrap inference after BIC selection | Restricting model selection within the bootstrap | Sparse capture data | The stepwise method | Bootstrap inference after stepwise selection | Fitting a specified model | The Bayesian thresholding method | Which method should I use? | Looking inside the result | References

Introduction | The Bayesian-threshold procedure | Bayesian thresholding | MCMC estimation | MCMC settings and posterior output | A basic analysis | Improper priors | Four-list interactions | Checking existence with improper interaction priors | Examples with improper priors | Summary | References

Introduction | BIC model selection | Inspecting the fit | Choosing the interaction order | Bootstrap inference | Practical bootstrap inference | Reading the confidence limits | Choosing the retained model set | Summary | References

Introduction | Running the reproduction code | Applications and published results | Table 2: Kosovo | Table 3: Korea | Figure 1: Five-list UK data | Figure 2: Five-list New Orleans data | Further investigations from the 2024 paper | Table 4: Location of the bootstrap BIC minimum | Table 5: Degree-2 ranking for Kosovo | Figure 3: Degree-1 and degree-2 rankings | Figure 4: Rank of the bootstrap-optimal model | Table 6: Downhill model search | Summary | References

Introduction | Sparse capture data and extended maximum likelihood | Existence and identifiability | The stepwise model-selection procedure | Allowing higher-order interactions | Bootstrap inference | Applications and published results | New Orleans: Section 4.1 | Five-list New Orleans data | Western site data: Section 4.2 | Simulation studies | Comparison with the BIC approach: Section 3.4 | Choosing the threshold: Section 4.3 | Relation to the current package | References