Convergence (bristlecone) | bristlecone

bristlecone

Convergence Module

Namespace: Bristlecone
Assembly: Bristlecone.dll
Parent Module: Diagnostics

Convergence diagnostics for monte-carlo markov chain (MCMC) analyses.

Types

Type	Description
ConvergenceStatistic	A per-parameter convergence statistic. The statistic used is given in `StatisticName`.

Functions and values

Function or value	Description
`gelmanRubin nMostRecent subjectId hypothesisId result` Full Usage: `gelmanRubin nMostRecent subjectId hypothesisId result` Parameters: nMostRecent : `int` - How many recent iterations to use from the trace. subjectId : `string` - The subject identifier hypothesisId : `string` - The hypothesis identifier result : `ResultSet<'subject, 'hypothesis>` - A result set (of 1 .. many results) for a particular subject and hypothesis Returns: `ConvergenceStatistic seq option` If more than one replicate, the R-hat convergence statistic across replicates	Calculate the Gelman-Rubin statistic for each parameter in the given `ResultSet`. The statistic tends downwards to one, with one indicating perfect convergence between all chains. nMostRecent : `int` How many recent iterations to use from the trace. subjectId : `string` The subject identifier hypothesisId : `string` The hypothesis identifier result : `ResultSet<'subject, 'hypothesis>` A result set (of 1 .. many results) for a particular subject and hypothesis Returns: `ConvergenceStatistic seq option` If more than one replicate, the R-hat convergence statistic across replicates
`gelmanRubinAll nMostRecent subject hypothesis results` Full Usage: `gelmanRubinAll nMostRecent subject hypothesis results` Parameters: nMostRecent : `int` - How many recent iterations to use from the trace. subject : `'subject -> string` - A function to retrieve a subject ID from a subject hypothesis : `'hypothesis -> string` - A function to retrieve a hypothesis ID from a hypothesis results : `ResultSet<'subject, 'hypothesis> seq` Returns: `ConvergenceStatistic seq` If more than one replicate, the R-hat convergence statistic across replicates	Calculate the Gelman-Rubin statistic for each parameter in all of the given `ResultSet`. The statistic tends downwards to one, with one indicating perfect convergence between all chains. nMostRecent : `int` How many recent iterations to use from the trace. subject : `'subject -> string` A function to retrieve a subject ID from a subject hypothesis : `'hypothesis -> string` A function to retrieve a hypothesis ID from a hypothesis results : `ResultSet<'subject, 'hypothesis> seq` Returns: `ConvergenceStatistic seq` If more than one replicate, the R-hat convergence statistic across replicates