Class

com.github.jonnylaw.model

ParticleMetropolis

Related Doc: package model

Permalink

case class ParticleMetropolis(logLikelihood: (Parameters) ⇒ LogLikelihood, initialParams: Parameters, proposal: (Parameters) ⇒ Rand[Parameters]) extends MetropolisHastings with Product with Serializable

Implementation of the particle metropolis algorithm without a properly specified prior distribution

logLikelihood

a function from parameters to LogLikelihood

initialParams

the starting parameters for the metropolis algorithm

proposal

a SYMMETRIC proposal distribution for the metropolis algorithm (eg. Gaussian)

Linear Supertypes
Serializable, Serializable, Product, Equals, MetropolisHastings, AnyRef, Any
Ordering
  1. Alphabetic
  2. By Inheritance
Inherited
  1. ParticleMetropolis
  2. Serializable
  3. Serializable
  4. Product
  5. Equals
  6. MetropolisHastings
  7. AnyRef
  8. Any
  1. Hide All
  2. Show All
Visibility
  1. Public
  2. All

Instance Constructors

  1. new ParticleMetropolis(logLikelihood: (Parameters) ⇒ LogLikelihood, initialParams: Parameters, proposal: (Parameters) ⇒ Rand[Parameters])

    Permalink

    logLikelihood

    a function from parameters to LogLikelihood

    initialParams

    the starting parameters for the metropolis algorithm

    proposal

    a SYMMETRIC proposal distribution for the metropolis algorithm (eg. Gaussian)

Value Members

  1. final def !=(arg0: Any): Boolean

    Permalink
    Definition Classes
    AnyRef → Any
  2. final def ##(): Int

    Permalink
    Definition Classes
    AnyRef → Any
  3. final def ==(arg0: Any): Boolean

    Permalink
    Definition Classes
    AnyRef → Any
  4. final def asInstanceOf[T0]: T0

    Permalink
    Definition Classes
    Any
  5. def clone(): AnyRef

    Permalink
    Attributes
    protected[java.lang]
    Definition Classes
    AnyRef
    Annotations
    @throws( ... )
  6. final def eq(arg0: AnyRef): Boolean

    Permalink
    Definition Classes
    AnyRef
  7. def finalize(): Unit

    Permalink
    Attributes
    protected[java.lang]
    Definition Classes
    AnyRef
    Annotations
    @throws( classOf[java.lang.Throwable] )
  8. final def getClass(): Class[_]

    Permalink
    Definition Classes
    AnyRef → Any
  9. val initialParams: Parameters

    Permalink

    the starting parameters for the metropolis algorithm

    the starting parameters for the metropolis algorithm

    Definition Classes
    ParticleMetropolisMetropolisHastings
  10. final def isInstanceOf[T0]: Boolean

    Permalink
    Definition Classes
    Any
  11. def iters: Process[MetropState]

    Permalink

    Use the Breeze Markov Chain to generate a process of MetropState Calling .sample(n) on this will create a single site metropolis hastings, proposing parameters only from the initial supplied parameter values

    Use the Breeze Markov Chain to generate a process of MetropState Calling .sample(n) on this will create a single site metropolis hastings, proposing parameters only from the initial supplied parameter values

    Definition Classes
    MetropolisHastings
  12. def itersAkka: Source[MetropState, Any]

    Permalink

    Generates an akka stream of MetropState, containing the current parameters, count of accepted moves and the current pseudo marginal log-likelihood Unfortunately for an unknown reason this isn't working

    Generates an akka stream of MetropState, containing the current parameters, count of accepted moves and the current pseudo marginal log-likelihood Unfortunately for an unknown reason this isn't working

    Definition Classes
    MetropolisHastings
  13. def itersSeq(n: Int): Seq[MetropState]

    Permalink

    Returns iterations from the MCMC algorithm in a vector using sampleStep, sampleStep

    Returns iterations from the MCMC algorithm in a vector using sampleStep, sampleStep

    Definition Classes
    MetropolisHastings
  14. def itersStream: Source[MetropState, Any]

    Permalink

    Use the same step for iterations in a stream

    Use the same step for iterations in a stream

    Definition Classes
    MetropolisHastings
  15. val logLikelihood: (Parameters) ⇒ LogLikelihood

    Permalink

    a function from parameters to LogLikelihood

    a function from parameters to LogLikelihood

    Definition Classes
    ParticleMetropolisMetropolisHastings
  16. def logTransition(from: Parameters, to: Parameters): LogLikelihood

    Permalink

    Definition of the log-transition, used when calculating the acceptance ratio This is the probability of moving between parameters according to the proposal distribution Note: When using a symmetric proposal distribution (eg.

    Definition of the log-transition, used when calculating the acceptance ratio This is the probability of moving between parameters according to the proposal distribution Note: When using a symmetric proposal distribution (eg. Normal) this cancels in the acceptance ratio

    from

    the previous parameter value

    to

    the proposed parameter value

    Definition Classes
    ParticleMetropolisMetropolisHastings
  17. def mhStep: (MetropState) ⇒ MetropState

    Permalink

    Generic metropolis-hastings step, which can be used with the usual acceptance ratio or simplified to the metropolis ratio by specifying the log-transition of the parameters to be zero

    Generic metropolis-hastings step, which can be used with the usual acceptance ratio or simplified to the metropolis ratio by specifying the log-transition of the parameters to be zero

    Definition Classes
    MetropolisHastings
  18. def mhStepRand: (MetropState) ⇒ Rand[MetropState]

    Permalink

    A single step of the metropolis hastings algorithm to be used with breeze implementation of Markov Chain.

    A single step of the metropolis hastings algorithm to be used with breeze implementation of Markov Chain. This is a slight alteration to the implementation in breeze, here MetropState holds on to the previous calculated pseudo marginal log-likelihood value so we don't need to run the previous particle filter again each iteration

    Definition Classes
    MetropolisHastings
  19. final def ne(arg0: AnyRef): Boolean

    Permalink
    Definition Classes
    AnyRef
  20. final def notify(): Unit

    Permalink
    Definition Classes
    AnyRef
  21. final def notifyAll(): Unit

    Permalink
    Definition Classes
    AnyRef
  22. def paramsAkka: Source[Parameters, Any]

    Permalink

    Return an akka stream of the parameters

    Return an akka stream of the parameters

    Definition Classes
    MetropolisHastings
  23. def prior: ContinuousDistr[Parameters]

    Permalink

    Prior distribution for the parameters, with default implementation

    Prior distribution for the parameters, with default implementation

    Definition Classes
    MetropolisHastings
  24. val proposal: (Parameters) ⇒ Rand[Parameters]

    Permalink

    a SYMMETRIC proposal distribution for the metropolis algorithm (eg.

    a SYMMETRIC proposal distribution for the metropolis algorithm (eg. Gaussian)

    Definition Classes
    ParticleMetropolisMetropolisHastings
  25. final def synchronized[T0](arg0: ⇒ T0): T0

    Permalink
    Definition Classes
    AnyRef
  26. final def wait(): Unit

    Permalink
    Definition Classes
    AnyRef
    Annotations
    @throws( ... )
  27. final def wait(arg0: Long, arg1: Int): Unit

    Permalink
    Definition Classes
    AnyRef
    Annotations
    @throws( ... )
  28. final def wait(arg0: Long): Unit

    Permalink
    Definition Classes
    AnyRef
    Annotations
    @throws( ... )

Inherited from Serializable

Inherited from Serializable

Inherited from Product

Inherited from Equals

Inherited from MetropolisHastings

Inherited from AnyRef

Inherited from Any

Ungrouped