Merger using importance sampling with mixture proposal density. More...
#include <merger.h>
Public Member Functions | |
| virtual string | to_string () |
| This method returns a basic info about the current instance. | |
| virtual void | to_setting (Setting &set) const |
| This method save all the instance properties into the Setting structure. | |
Constructors | |
| merger_mix () | |
| merger_mix (const Array< shared_ptr< mpdf > > &S) | |
| void | set_sources (const Array< shared_ptr< mpdf > > &S) |
| Set sources and prepare all internal structures. | |
| void | set_parameters (int Ncoms0=DFLT_Ncoms, double effss_coef0=DFLT_effss_coef) |
| Set internal parameters used in approximation. | |
Mathematical operations | |
| void | merge () |
| Merge values using mixture approximation. | |
| vec | sample () const |
| sample from the approximating mixture | |
| double | evallog (const vec &dt) const |
| loglikelihood computed on mixture models | |
Access functions | |
| MixEF & | _Mix () |
| Access function. | |
| emix * | proposal () |
| Access function. | |
| void | from_setting (const Setting &set) |
| from_settings | |
Constructors | |
| void | set_support (rectangular_support &Sup) |
| Set support points from rectangular grid. | |
| void | set_support (discrete_support &Sup) |
| Set support points from dicrete grid. | |
| void | set_support (const epdf &overall, int N) |
| Set support points from a pdf by drawing N samples. | |
| void | set_debug_file (const string fname) |
| set debug file | |
| void | set_method (MERGER_METHOD MTH=DFLT_METHOD, double beta0=DFLT_beta) |
| Set internal parameters used in approximation. | |
Mathematical operations | |
| vec | merge_points (mat &lW) |
| Merge log-likelihood values in points using method specified by parameter METHOD. | |
| vec | mean () const |
| weight w is a | |
| mat | covariance () const |
| vec | variance () const |
| return expected variance (not covariance!) | |
Access to attributes | |
| eEmp & | _Smp () |
| Access function. | |
| void | validate () |
| This method TODO. | |
Constructors | |
Construction of each epdf should support two types of constructors:
The following constructors should be supported for convenience:
All internal data structures are constructed as empty. Their values (including sizes) will be set by method | |
| void | set_parameters (int dim0) |
Matematical Operations | |
| virtual mat | sample_m (int N) const |
Returns N samples, from density . | |
| virtual vec | evallog_m (const mat &Val) const |
Compute log-probability of multiple values argument val. | |
| virtual vec | evallog_m (const Array< vec > &Avec) const |
Compute log-probability of multiple values argument val. | |
| virtual shared_ptr< mpdf > | condition (const RV &rv) const |
| Return conditional density on the given RV, the remaining rvs will be in conditioning. | |
| virtual shared_ptr< epdf > | marginal (const RV &rv) const |
| Return marginal density on the given RV, the remainig rvs are intergrated out. | |
| virtual void | qbounds (vec &lb, vec &ub, double percentage=0.95) const |
Lower and upper bounds of percentage % quantile, returns mean-2*sigma as default. | |
Connection to other classes | |
| void | set_rv (const RV &rv0) |
| Name its rv. | |
| bool | isnamed () const |
| True if rv is assigned. | |
| const RV & | _rv () const |
| Return name (fails when isnamed is false). | |
Access to attributes | |
| int | dimension () const |
| Size of the random variable. | |
Protected Attributes | |
| MixEF | Mix |
| Internal mixture of EF models. | |
| int | Ncoms |
| Number of components in a mixture. | |
| double | effss_coef |
| coefficient of resampling [0,1] | |
| int | stop_niter |
| stop after niter iterations | |
| Array< shared_ptr< mpdf > > | mpdfs |
| Elements of composition. | |
| Array< datalink_m2e * > | dls |
| Data link for each mpdf in mpdfs. | |
| Array< RV > | rvzs |
Array of rvs that are not modelled by mpdfs at all, . | |
| Array< datalink_m2e * > | zdls |
Data Links for extension . | |
| int | Npoints |
| number of support points | |
| int | Nsources |
| number of sources | |
| MERGER_METHOD | METHOD |
| switch of the methoh used for merging | |
| double | beta |
| Prior on the log-normal merging model. | |
| eEmp | eSmp |
| Projection to empirical density (could also be piece-wise linear). | |
| bool | DBG |
| debug or not debug | |
| it_file * | dbg_file |
| debugging file | |
| int | dim |
| dimension of the random variable | |
| RV | rv |
| Description of the random variable. | |
Static Protected Attributes | |
| static const int | DFLT_Ncoms = 10 |
| default value for Ncoms | |
| static const double | DFLT_effss_coef = 0.5 |
| default value for efss_coef; | |
| static const MERGER_METHOD | DFLT_METHOD = LOGNORMAL |
| Default for METHOD. | |
| static const double | DFLT_beta = 1.2 |
| default for beta | |
Merger using importance sampling with mixture proposal density.
| vec bdm::merger_base::mean | ( | ) | const [inline, virtual, inherited] |
weight w is a
sample from merged density
Reimplemented from bdm::epdf.
References bdm::eEmp::_samples(), bdm::eEmp::_w(), bdm::epdf::dim, bdm::merger_base::eSmp, and bdm::merger_base::Npoints.
Referenced by bdm::merger_base::variance().
1.6.1