Inference Using Simulation
Create or augment a list of simulated distributions of summary statist...
Create or augment a list of simulated distributions of summary statist...
Check linear dependencies among raw summary statistics
Compute confidence intervals by (profile) summary likelihood
Specificying arbitrary constraints on parameters
Wrapper to generate projection functions for all parameters
Internal S4 classes.
Workflow for primitive method, with projections
Workflow for primitive method, without projections
Workflow for method with reference table
Summary, print and logLik methods for Infusion results.
Refine summary likelihood profile in focal parameter values
Backward-compatible extractor from summary-likelihood objects
Control of number of components in Gaussian mixture modelling
Workflow design
Assessing goodness of fit of inference using simulation
Discrete probability masses and NA/NaN/Inf in distributions of summary...
Infer log Likelihoods using simulated distributions of summary statist...
Infer a (summary) likelihood surface from a simulation table
Infer a (summary) likelihood or tail probability surface from inferred...
Internal Infusion Functions
Inference using simulation
Define starting points in parameter space.
Modeling and predicting latent variables
Control of MAF design and training
Multivariate histogram
Maximum likelihood from an inferred likelihood surface
Infusion options settings
Diagnostic plots for projections
Plot fit objects
Plot likelihood profiles
Evaluate log-likelihood for given parameters
Compute profile summary likelihood
Learn a projection method for statistics and apply it
Refine estimates iteratively
Conversion to new parameter spaces
Sample the parameter space
Save or load MAF Python objects
Simulate method for an SLik_j object.
Summary likelihood ratio tests
Model density evaluation for given data and parameters
Updating an 'SLik_j' object for new data
Implements functions for simulation-based inference. In particular, implements functions to perform likelihood inference from data summaries whose distributions are simulated, as first described in Rousset et al. (2017) <doi:10.1111/1755-0998.12627>. The package implements more advanced methods described in Rousset et al. (2025) <doi:10.1101/2024.09.30.615940>.