MPF PF libPF.h MPF::mpfepdf class BM_T BM_T BM_T * BM_T* MPF< BM_T >::Bms[10000] [10000] Bms mpfepdf mpfepdf MPF< BM_T >::jest jest estimate joining PF.est with conditional double double MPF< BM_T >::SSAT SSAT MPF< BM_T >::MPF (const RV &rvlin, const RV &rvpf, mpdf &par0, mpdf &obs0, int n, const BM_T &BMcond0) MPF const RV & rvlin const RV & rvpf mpdf & par0 mpdf & obs0 int n const BM_T & BMcond0 Default constructor. RV::add BM::rv MPF< BM_T >::~MPF () ~MPF void void MPF< BM_T >::bayes (const vec &dt) bayes bayes const vec & dt Incremental Bayes rule. dt vector of input data PF::_samples PF::_w PF::est PF::n PF::par eEmp::resample mpdf::samplecond epdf & epdf& MPF< BM_T >::_epdf () _epdf _epdf Returns a pointer to the epdf representing posterior density on parameters. Use with care! void void MPF< BM_T >::set_est (const epdf &epdf0) set_est set_est const epdf & epdf0 Set postrior of rvc to samples from epdf0. Statistics of Bms are not re-computed! Use only for initialization! PF::_samples PF::n PF::set_est void void BM::bayes (mat Dt) bayes mat Dt Batch Bayes rule (columns of Dt are observations). const RV & const RV& BM::_rv () const _rv access function BM::rv double double BM::_ll () const _ll access function BM::ll int int PF::n n number of particles; MPF< BM_T >::bayes PF::bayes MPF< BM_T >::set_est PF::set_est eEmp eEmp PF::est est posterior density MPF< BM_T >::bayes PF::bayes vec & vec& PF::_w _w pointer into eEmp MPF< BM_T >::bayes PF::bayes Array< vec > & Array<vec>& PF::_samples _samples pointer into eEmp MPF< BM_T >::bayes PF::bayes MPF< BM_T >::set_est PF::set_est mpdf & mpdf& PF::par par Parameter evolution model. MPF< BM_T >::bayes PF::bayes mpdf & mpdf& PF::obs obs Observation model. PF::bayes RV RV BM::rv rv Random variable of the posterior. BM::_rv MPF< BM_T >::MPF EKFfull::set_parameters ARX::structure_est double double BM::ll ll Logarithm of marginalized data likelihood. BM::_ll EKFfixed::bayes EKF< sq_T >::bayes Kalman< sq_T >::bayes EKFCh::bayes KalmanCh::bayes EKFfull::bayes ARX::bayes bool bool BM::evalll evalll If true, the filter will compute likelihood of the data record and store it in ll . Set to false if you want to save time. EKFfixed::bayes EKF< sq_T >::bayes Kalman< sq_T >::bayes EKFCh::bayes KalmanCh::bayes EKFfull::bayes ARX::bayes Marginalized Particle filter. Trivial version: proposal = parameter evolution, observation model is not used. (it is assumed to be part of BM). BM MPF< BM_T > PF BM_T epdf rv BM rv MPF< BM_T > Bms jest PF obs par est eEmp RV mpdf rv rvc ep MPF< BM_T >::mpfepdf E MPF_epdf MPF_ll MPF_rv MPF_samples MPF_w MPFbayes MPFbayes MPFBM MPFBms MPFest MPFevalll MPFjest MPFll MPFMPF MPFn MPFobs MPFpar MPFPF MPFrv MPFset_est MPFSSAT MPF~BM MPF~MPF