Stan  2.10.0
probability, sampling & optimization
Classes | Functions
stan::mcmc Namespace Reference

Markov chain Monte Carlo samplers. More...

Classes

class  adapt_dense_e_nuts
 The No-U-Turn sampler (NUTS) with multinomial sampling with a Gaussian-Euclidean disintegration and adaptive dense metric and adaptive step size. More...
 
class  adapt_dense_e_nuts_classic
 
class  adapt_dense_e_static_hmc
 Hamiltonian Monte Carlo implementation using the endpoint of trajectories with a static integration time with a Gaussian-Euclidean disintegration and adative dense metric and adaptive step size. More...
 
class  adapt_dense_e_static_uniform
 Hamiltonian Monte Carlo implementation that uniformly samples from trajectories with a static integration time with a Gaussian-Euclidean disintegration and adaptive dense metric and adaptive step size. More...
 
class  adapt_dense_e_xhmc
 Exhausive Hamiltonian Monte Carlo (XHMC) with multinomial sampling with a Gaussian-Euclidean disintegration and adaptive dense metric and adaptive step size. More...
 
class  adapt_diag_e_nuts
 The No-U-Turn sampler (NUTS) with multinomial sampling with a Gaussian-Euclidean disintegration and adaptive diagonal metric and adaptive step size. More...
 
class  adapt_diag_e_nuts_classic
 
class  adapt_diag_e_static_hmc
 Hamiltonian Monte Carlo implementation using the endpoint of trajectories with a static integration time with a Gaussian-Euclidean disintegration and adaptive diagonal metric and adaptive step size. More...
 
class  adapt_diag_e_static_uniform
 Hamiltonian Monte Carlo implementation that uniformly samples from trajectories with a static integration time with a Gaussian-Euclidean disintegration and adaptive diagonal metric and adaptive step size. More...
 
class  adapt_diag_e_xhmc
 Exhausive Hamiltonian Monte Carlo (XHMC) with multinomial sampling with a Gaussian-Euclidean disintegration and adaptive diagonal metric and adaptive step size. More...
 
class  adapt_softabs_nuts
 The No-U-Turn sampler (NUTS) with multinomial sampling with a Gaussian-Riemannian disintegration and SoftAbs metric and adaptive step size. More...
 
class  adapt_softabs_static_hmc
 Hamiltonian Monte Carlo implementation using the endpoint of trajectories with a static integration time with a Gaussian-Riemannian disintegration and SoftAbs metric and adaptive step size. More...
 
class  adapt_softabs_static_uniform
 Hamiltonian Monte Carlo implementation that uniformly samples from trajectories with a static integration time with a Gaussian-Riemannian disintegration and SoftAbs metric and adaptive step size. More...
 
class  adapt_softabs_xhmc
 Exhausive Hamiltonian Monte Carlo (XHMC) with multinomial sampling with a Gaussian-Riemannian disintegration and SoftAbs metric and adaptive step size. More...
 
class  adapt_unit_e_nuts
 The No-U-Turn sampler (NUTS) with multinomial sampling with a Gaussian-Euclidean disintegration and unit metric and adaptive step size. More...
 
class  adapt_unit_e_nuts_classic
 
class  adapt_unit_e_static_hmc
 Hamiltonian Monte Carlo implementation using the endpoint of trajectories with a static integration time with a Gaussian-Euclidean disintegration and unit metric and adaptive step size. More...
 
class  adapt_unit_e_static_uniform
 Hamiltonian Monte Carlo implementation that uniformly samples from trajectories with a static integration time with a Gaussian-Euclidean disintegration and unit metric and adaptive step size. More...
 
class  adapt_unit_e_xhmc
 Exhausive Hamiltonian Monte Carlo (XHMC) with multinomial sampling with a Gaussian-Euclidean disintegration and unit metric and adaptive step size. More...
 
class  base_adaptation
 
class  base_adapter
 
class  base_hamiltonian
 
class  base_hmc
 
class  base_integrator
 
class  base_leapfrog
 
class  base_mcmc
 
class  base_nuts
 The No-U-Turn sampler (NUTS) with multinomial sampling. More...
 
class  base_nuts_classic
 
class  base_static_hmc
 Hamiltonian Monte Carlo implementation using the endpoint of trajectories with a static integration time. More...
 
class  base_static_uniform
 Hamiltonian Monte Carlo implementation that uniformly samples from trajectories with a static integration time. More...
 
class  base_xhmc
 Exhaustive Hamiltonian Monte Carlo (XHMC) with multinomial sampling. More...
 
class  chains
 An mcmc::chains object stores parameter names and dimensionalities along with samples from multiple chains. More...
 
class  covar_adaptation
 
class  dense_e_metric
 
class  dense_e_nuts
 The No-U-Turn sampler (NUTS) with multinomial sampling with a Gaussian-Euclidean disintegration and dense metric. More...
 
class  dense_e_nuts_classic
 
class  dense_e_point
 Point in a phase space with a base Euclidean manifold with dense metric. More...
 
class  dense_e_static_hmc
 Hamiltonian Monte Carlo implementation using the endpoint of trajectories with a static integration time with a Gaussian-Euclidean disintegration and dense metric. More...
 
class  dense_e_static_uniform
 Hamiltonian Monte Carlo implementation that uniformly samples from trajectories with a static integration time with a Gaussian-Euclidean disintegration and dense metric. More...
 
class  dense_e_xhmc
 Exhausive Hamiltonian Monte Carlo (XHMC) with multinomial sampling with a Gaussian-Euclidean disintegration and dense metric. More...
 
class  diag_e_metric
 
class  diag_e_nuts
 The No-U-Turn sampler (NUTS) with multinomial sampling with a Gaussian-Euclidean disintegration and diagonal metric. More...
 
class  diag_e_nuts_classic
 
class  diag_e_point
 Point in a phase space with a base Euclidean manifold with diagonal metric. More...
 
class  diag_e_static_hmc
 Hamiltonian Monte Carlo implementation using the endpoint of trajectories with a static integration time with a Gaussian-Euclidean disintegration and diagonal metric. More...
 
class  diag_e_static_uniform
 Hamiltonian Monte Carlo implementation that uniformly samples from trajectories with a static integration time with a Gaussian-Euclidean disintegration and diagonal metric. More...
 
class  diag_e_xhmc
 Exhausive Hamiltonian Monte Carlo (XHMC) with multinomial sampling with a Gaussian-Euclidean disintegration and diagonal metric. More...
 
class  expl_leapfrog
 
class  fixed_param_sampler
 
class  impl_leapfrog
 
struct  nuts_util
 
class  ps_point
 Point in a generic phase space. More...
 
class  sample
 
struct  softabs_fun
 
class  softabs_metric
 
class  softabs_nuts
 The No-U-Turn sampler (NUTS) with multinomial sampling with a Gaussian-Riemannian disintegration and SoftAbs metric. More...
 
class  softabs_point
 Point in a phase space with a base Riemannian manifold with SoftAbs metric. More...
 
class  softabs_static_hmc
 Hamiltonian Monte Carlo implementation using the endpoint of trajectories with a static integration time with a Gaussian-Riemannian disintegration and SoftAbs metric. More...
 
class  softabs_static_uniform
 Hamiltonian Monte Carlo implementation that uniformly samples from trajectories with a static integration time with a Gaussian-Riemannian disintegration and SoftAbs metric. More...
 
class  softabs_xhmc
 Exhausive Hamiltonian Monte Carlo (XHMC) with multinomial sampling with a Gaussian-Riemannian disintegration and SoftAbs metric. More...
 
class  stepsize_adaptation
 
class  stepsize_adapter
 
class  stepsize_covar_adapter
 
class  stepsize_var_adapter
 
class  unit_e_metric
 
class  unit_e_nuts
 The No-U-Turn sampler (NUTS) with multinomial sampling with a Gaussian-Euclidean disintegration and unit metric. More...
 
class  unit_e_nuts_classic
 
class  unit_e_point
 Point in a phase space with a base Euclidean manifold with unit metric. More...
 
class  unit_e_static_hmc
 Hamiltonian Monte Carlo implementation using the endpoint of trajectories with a static integration time with a Gaussian-Euclidean disintegration and unit metric. More...
 
class  unit_e_static_uniform
 Hamiltonian Monte Carlo implementation that uniformly samples from trajectories with a static integration time with a Gaussian-Euclidean disintegration and unit metric. More...
 
class  unit_e_xhmc
 Exhausive Hamiltonian Monte Carlo (XHMC) with multinomial sampling with a Gaussian-Euclidean disintegration and unit metric. More...
 
class  var_adaptation
 
class  windowed_adaptation
 

Functions

void write_metric (stan::interface_callbacks::writer::base_writer &writer)
 
void stable_sum (double a1, double log_w1, double a2, double log_w2, double &sum_a, double &log_sum_w)
 a1 and a2 are running averages of the form $ a1 = ( \sum_{n \in N1} w_{n} f_{n} ) / ( \sum_{n \in N1} w_{n} ) $ $ a2 = ( \sum_{n \in N2} w_{n} f_{n} ) / ( \sum_{n \in N2} w_{n} ) $ and the weights are the respective normalizing constants $ w1 = \sum_{n \in N1} w_{n} $ $ w2 = \sum_{n \in N2} w_{n}. $ More...
 

Detailed Description

Markov chain Monte Carlo samplers.

Function Documentation

void stan::mcmc::stable_sum ( double  a1,
double  log_w1,
double  a2,
double  log_w2,
double &  sum_a,
double &  log_sum_w 
)

a1 and a2 are running averages of the form $ a1 = ( \sum_{n \in N1} w_{n} f_{n} ) / ( \sum_{n \in N1} w_{n} ) $ $ a2 = ( \sum_{n \in N2} w_{n} f_{n} ) / ( \sum_{n \in N2} w_{n} ) $ and the weights are the respective normalizing constants $ w1 = \sum_{n \in N1} w_{n} $ $ w2 = \sum_{n \in N2} w_{n}. $

This function returns the pooled average $ sum_a = ( \sum_{n \in N1 \cup N2} w_{n} f_{n} ) / ( \sum_{n \in N1 \cup N2} w_{n} ) $ and the pooled weights $ log_sum_w = log(w1 + w2). $

Parameters
a1First running average, f1 / w1
log_w1Log of first summed weight
a2Second running average
log_w2Log of second summed weight
sum_aAverage of input running averages
log_sum_wLog of summed input weights

Definition at line 40 of file base_xhmc.hpp.

void stan::mcmc::write_metric ( stan::interface_callbacks::writer::base_writer writer)

Definition at line 18 of file unit_e_point.hpp.


     [ Stan Home Page ] © 2011–2016, Stan Development Team.