Method ao/mcmc


  MCMC estimates paramters using a Monte Carlo Markov Chain.
 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
 
  DESCRIPTION: MCMC estimate the parameters of a given model given
               inputs, outputs and noise using a Metropolis algorithm.
               It handles (1 input / 1 output) systems, (2 input / 1 output) systems,
               and (2 input / 2 output) systems.
 
  CALL:        b = mcmc(in,out,pl)
 
  INPUTS:      out     - analysis objects with measured outputs
 
               pl      - parameter list
 
  OUTPUTS:     b   - pest object contatining estimate information
 
  Parameters Description
 
  VERSION:    $Id: mcmc.m,v 1.25 2011/12/19 14:45:10 mauro Exp $
 
  References:  "Catching supermassive black holes binaries without a net"
               N.J. Cornish, E.K. Porter, Phys.Rev.D 75, 021301, 2007
 
  TODO: multiple chain option not implemented yet
        metropolis/hastings not implemented
 
 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
Method Details
Access public
Defining Class ao
Sealed 0
Static 0

Parameter Description

Default

no description
Key Default Value Options Description
N 1000 none number of samples of the chain.
COV 0.0001 none covariance of the gaussian jumping distribution.
FITPARAMS [] none A cell array of evaluated parameters.
INPUT '' none A matrix array of input signals.
FREQUENCIES [] none Range of frequencies where the analysis is performed. If an array, only first and last are used
NOISE '' none An array of noise spectrum (PSD) used to compute the likelihood.
    MODEL '' none An array of models.
    SEARCH 1
    • 1
    • 0
    Set to true to use bigger jumps in parameter space during annealing and cool down.
    SIMPLEX 1
    • 1
    • 0
    Set to true to perform a simplex search to find the starting parameters of the MCMC chain.
    HEAT 1 none The heat index flattening likelihood surface during annealing.
    TC [] none An array of two values setting the initial and final value for the cooling down.
    X0 [] none The proposed initial values.
    JUMPS [] none An array of four numbers setting the rescaling of the covariance matrix during the search phase.
    PLOT [] none Select indexes of the parameters to be plotted.
    DEBUG 0
    • 0
    • 1
    Set to true to get debug information of the MCMC process.
    PRIOR '' none Mean, sigma and normalization factor for priors. Still under test
    ANNEAL 'simul'
    • 'simul'
    • 'thermo'
    • 'simple'
    Choose type of annealing during sampling. Default value is simulated annealing. Choose "thermo" for annealing with a thermostat. SNR is computed and if it is larger than a fixed value SNR0 (provided also in the plist), then the chains are heated by a factor of (SNR(1)/SNR0)^2. Choosing "simple" the deviation of the loglikelihood of every 10 points in the chains is stored. If this deviation is larger or smaller than two fixed values the chains are cooled or heated respectively.
    SNR0 200 none Fixed value for thermostated annealing.
    DELTAL [100 600 2 3] none Deviation of Loglikelihood for 10 points of the chains. Used forthe "simple" choice of annealing with a thermostat.
    back to top back to top

    Some information of the method ao/mcmc are listed below:
    Class name ao
    Method name mcmc
    Category Signal Processing
    CVS Version $Id: mcmc.m,v 1.25 2011/12/19 14:45:10 mauro Exp $
    Min input args 1
    Max input args -1
    Min output args 1
    Max output args -1




    ©LTP Team