Analyzers

Analyzers observe the simulation each timestep and store derived results without changing dynamics. STIsim ships analyzers for coinfection, sex-work transmission, and network/partnership structure, on top of Starsim’s analyzer framework. Attach them via the analyzers=[...] argument to sti.Sim.

sim = sti.Sim(
    diseases=['hiv', 'syphilis'],
    analyzers=[sti.coinfection_stats(), sti.PartnershipFormationAnalyzer()],
)

Available analyzers

Analyzer	What it tracks
sti.coinfection_stats	Coinfection statistics for two diseases (conditional prevalences).
sti.sw_stats	New infections and transmissions among sex workers and their clients.
sti.art_coverage	ART coverage (number and proportion) by sex and age bin.
sti.RelationshipDurations	Relationship duration distributions in a StructuredSexual network.
sti.NetworkDegree	Lifetime partner-count distributions.
sti.TimeBetweenRelationships	Gaps between successive relationships.
sti.partner_age_diff	Age differences between sexual partners.
sti.DebutAge	Proportion of agents sexually debuted by age.
sti.PartnershipFormationAnalyzer	Partnership formation per network, gender, and age bin. Relies on networks recording the timestep each relationship ends.

The grouped-result analyzers build on result_grouper, which provides conditional probability utilities for stratified results.

Stub — expand with usage examples for each analyzer, the result keys they write, and how to plot them. See the API reference for analyzers and the Results tutorial for custom analysis patterns.