1 Data | 1.1 The randomREH3 dataset | 2 Step 1 — Processing with remify() | 2.1 The remify_durem object | 3 Step 2 — Statistics with remstats() | 3.1 Active-state effect functions | 3.2 Computing statistics | 3.3 Event types and consider_type | 4 Step 3 — Estimation with remstimate() | 4.1 Fitting the model | 4.2 Coefficients and summary | 4.3 Diagnostics | 4.4 Model comparison | 5 Additional topics | 5.1 Duration based event weights | 5.2 Right-censored events | 5.3 Simultaneous events and boundary coincidences | 5.3 Directed end process | References

1 Data | 2 Fitting the model | 2.1 Interpreting the output | 3 Selecting the number of classes | 3.1 Inspecting the selected model | 4 Diagnostics | 5 Comparison with MLE and GLMM | 6 Practical considerations | References

1 When to use a frailty model | 2 Data | 3 The standard model (no random effects) | 4 Random intercepts for actors | 4.1 Sender frailty | 4.2 Receiver frailty | 4.3 Crossed sender and receiver frailties | 5 Random slopes | 6 Model comparison | Both models have a preference towards the Receiver frailty model. | 7 Diagnostics | 8 Interpreting random effects | 9 Summary | References

1 When to use moving-window estimation | 2 Data | 3 Tie-oriented moving-window models | 3.1 Default (auto) windowing | 3.2 Manual, overlapping windows | 3.3 Reading the stability table | 4 Actor-oriented moving-window models | 4.1 Sender + receiver model | 4.2 Receiver-only actor model | 5 Diagnostics across windows | 5.1 Tie model | 5.2 Actor model | 6 Parallelization | 7 Current limitations | 8 Summary | References

1 Data | 2 The kitchen-sink model | 2.1 Unregularized MLE | 3 Bayesian regularization | 4 Frequentist regularization (elastic net) | 4.1 Comparing coefficients | 4.2 Alpha: lasso vs. ridge | 4.3 Lambda selection | 5 Diagnostics | 6 Summary | References

1 Data | 2 Tie-oriented models | 2.1 Basic model (interval timing) | Diagnostics | 2.2 Ordinal timing | 2.3 Risk set variations | Active risk set | Active-saturated risk set | Manual risk set | Extending the risk set by event type | 2.4 Typed events | consider_type = "ignore" | consider_type = "separate" | consider_type = "interact" | Estimation with typed events | Extending the risk set by type | 2.5 Exogenous statistics | Actor-level effects | Interactions between endogenous and exogenous effects | 2.6 Memory types | 2.7 Case-control sampling | 2.8 Model assessment and variable selection | Tuning the memory half-life | 3 Actor-oriented models | 3.1 Sender + receiver model | 3.2 Actor-oriented model with exogenous effects | 3.3 Actor-oriented diagnostics | 4 Bayesian estimation with HMC | 5 Putting it all together

1 When to use a frailty model | 2 Data | 3 The standard model (no random effects) | 4 Random intercepts for actors | 4.1 Sender frailty | 4.2 Receiver frailty | 4.3 Crossed sender and receiver frailties | 5 Random slopes | 6 Model comparison | Both models have a preference towards the Receiver frailty model. | 7 Diagnostics | 8 Interpreting random effects | 9 Summary | References

1 Data | 2 The kitchen-sink model | 2.1 Unregularized MLE | 3 Bayesian regularization | 4 Frequentist regularization (elastic net) | 4.1 Comparing coefficients | 4.2 Alpha: lasso vs. ridge | 4.3 Lambda selection | 5 Diagnostics | 6 Summary | References

1 When to use moving-window estimation | 2 Data | 3 Tie-oriented moving-window models | 3.1 Default (auto) windowing | 3.2 Manual, overlapping windows | 3.3 Reading the stability table | 4 Actor-oriented moving-window models | 4.1 Sender + receiver model | 4.2 Receiver-only actor model | 5 Diagnostics across windows | 5.1 Tie model | 5.2 Actor model | 6 Parallelization | 7 Current limitations | 8 Summary | References

1 Data | 1.1 The randomREH3 dataset | 2 Step 1 — Processing with remify() | 2.1 The remify_durem object | 3 Step 2 — Statistics with remstats() | 3.1 Active-state effect functions | 3.2 Computing statistics | 3.3 Event types and consider_type | 4 Step 3 — Estimation with remstimate() | 4.1 Fitting the model | 4.2 Coefficients and summary | 4.3 Diagnostics | 4.4 Model comparison | 5 Additional topics | 5.1 Duration based event weights | 5.2 Right-censored events | 5.3 Simultaneous events and boundary coincidences | 5.3 Directed end process | References

1 Data | 2 Fitting the model | 2.1 Interpreting the output | 3 Selecting the number of classes | 3.1 Inspecting the selected model | 4 Diagnostics | 5 Comparison with MLE and GLMM | 6 Practical considerations | References

1 Data | 2 Tie-oriented models | 2.1 Basic model (interval timing) | Diagnostics | 2.2 Ordinal timing | 2.3 Risk set variations | Active risk set | Active-saturated risk set | Manual risk set | Extending the risk set by event type | 2.4 Typed events | consider_type = "ignore" | consider_type = "separate" | consider_type = "interact" | Estimation with typed events | Extending the risk set by type | 2.5 Exogenous statistics | Actor-level effects | Interactions between endogenous and exogenous effects | 2.6 Memory types | 2.7 Case-control sampling | 2.8 Model assessment and variable selection | Tuning the memory half-life | 3 Actor-oriented models | 3.1 Sender + receiver model | 3.2 Actor-oriented model with exogenous effects | 3.3 Actor-oriented diagnostics | 4 Bayesian estimation with HMC | 5 Putting it all together

Estimation approaches: Frequentist and Bayesian | Let's get started (loading the remstimate package) | Modeling frameworks | Tie-Oriented Modeling framework | The likelihood function | A toy example on the tie oriented modeling framework | Estimating a model with remstimate() in 3 steps | Frequentist approach | Maximum Likelihood Estimation (MLE) | print( ) | summary( ) | Information Criteria | diagnostics( ) | plot( ) | Bayesian approach | Hamiltonian Monte Carlo (HMC) | Actor-Oriented Modeling framework | Sender activity rate model | Receiver choice model | A toy example on the actor oriented modeling framework | References

Definition of risk set | The full risk set | Visualizing the risk set | The active, active_saturated, and manual risk set | The active risk set | active dyads: those observed in randomREHsmall | The manual risk set | Specifying a manual risk set | The processed risk set

Aim | Input | edgelist | directed | ordinal | model | thin | actors | riskset | manual_riskset | event_type | origin | time.units | attach_riskset | riskset_decode | riskset_max_decode | event_covariates | ncores | Running the example | Output | M | E | N | C | D and activeD | intereventTime | edgelist_id | meta | ids | index | omit_dyad | riskset_info | Actor-oriented model output | Methods | print() and summary() | dim() | plot()

Introduction | Quick workflow example | Data | Tie-oriented model | Preparing the event history | Computing statistics | Memory | Event types and consider_type | Tie-oriented model with case-control sampling | Actor-oriented model