bdm Namespace Reference
Detailed Description
Space of basic BDM structures.
Classes | |
| class | ARX |
| Linear Autoregressive model with Gaussian noise. More... | |
| class | EKFful_unQR |
Extended Kalman filter with unknown Q and R. More... | |
| class | EKFCh_unQ |
Extended Kalman filter in Choleski form with unknown Q. More... | |
| class | EKFCh_cond |
Extended Kalman filter with unknown parameters in IM. More... | |
| class | KalmanFull |
| Basic Kalman filter with full matrices (education purpose only)! Will be deleted soon! More... | |
| class | Kalman |
| Kalman filter with covariance matrices in square root form. More... | |
| class | KalmanCh |
| Kalman filter in square root form. More... | |
| class | EKFfull |
| Extended Kalman Filter in full matrices. More... | |
| class | EKF |
| Extended Kalman Filter. More... | |
| class | EKFCh |
| Extended Kalman Filter in Square root. More... | |
| class | KFcondQR |
| Kalman Filter with conditional diagonal matrices R and Q. More... | |
| class | KFcondR |
| Kalman Filter with conditional diagonal matrices R and Q. More... | |
| class | PF |
| Trivial particle filter with proposal density equal to parameter evolution model. More... | |
| class | MPF |
| Marginalized Particle filter. More... | |
| class | merger |
| Function for general combination of pdfs. More... | |
| class | MixEF |
| Mixture of Exponential Family Densities. More... | |
| class | mratio |
| Class representing ratio of two densities which arise e.g. by applying the Bayes rule. It represents density in the form:
where | |
| class | emix |
| Mixture of epdfs. More... | |
| class | mprod |
| Chain rule decomposition of epdf. More... | |
| class | eprod |
| Product of independent epdfs. For dependent pdfs, use mprod. More... | |
| class | mmix |
| Mixture of mpdfs with constant weights, all mpdfs are of equal type. More... | |
| class | bdmroot |
| Root class of BDM objects. More... | |
| class | str |
| Structure of RV (used internally), i.e. expanded RVs. More... | |
| class | RV |
| Class representing variables, most often random variables. More... | |
| class | fnc |
Class representing function  of variable  represented by rv. More... | |
| class | epdf |
| Probability density function with numerical statistics, e.g. posterior density. More... | |
| class | mpdf |
| Conditional probability density, e.g. modeling some dependencies. More... | |
| class | datalink |
| DataLink is a connection between two data vectors Up and Down. More... | |
| class | datalink_m2e |
| data link between More... | |
| class | datalink_m2m |
| class | logger |
| Class for storing results (and semi-results) of an experiment. More... | |
| class | mepdf |
| Unconditional mpdf, allows using epdf in the role of mpdf. More... | |
| class | compositepdf |
| Abstract composition of pdfs, will be used for specific classes this abstract class is common to epdf and mpdf. More... | |
| class | DS |
| Abstract class for discrete-time sources of data. More... | |
| class | BM |
| Bayesian Model of a system, i.e. all uncertainty is modeled by probabilities. More... | |
| class | BMcond |
| Conditional Bayesian Filter. More... | |
| class | MemDS |
| Memory storage of off-line data column-wise. More... | |
| class | ArxDS |
| Generator of ARX data. More... | |
| class | stateDS |
| class | eEF |
| General conjugate exponential family posterior density. More... | |
| class | mEF |
| Exponential family model. More... | |
| class | BMEF |
| Estimator for Exponential family. More... | |
| class | enorm |
| Gaussian density with positive definite (decomposed) covariance matrix. More... | |
| class | egiw |
| Gauss-inverse-Wishart density stored in LD form. More... | |
| class | eDirich |
| Dirichlet posterior density. More... | |
| class | multiBM |
| Estimator for Multinomial density. More... | |
| class | egamma |
| Gamma posterior density. More... | |
| class | eigamma |
| Inverse-Gamma posterior density. More... | |
| class | euni |
| Uniform distributed density on a rectangular support. More... | |
| class | mlnorm |
| Normal distributed linear function with linear function of mean value;. More... | |
| class | mlstudent |
| class | mgamma |
| Gamma random walk. More... | |
| class | migamma |
| Inverse-Gamma random walk. More... | |
| class | mgamma_fix |
| Gamma random walk around a fixed point. More... | |
| class | migamma_fix |
| Inverse-Gamma random walk around a fixed point. More... | |
| class | eEmp |
| Weighted empirical density. More... | |
| class | constfn |
class representing function  , here rv is empty More... | |
| class | linfn |
Class representing function  . More... | |
| class | diffbifn |
Class representing a differentiable function of two variables  . More... | |
| class | bilinfn |
Class representing function  . More... | |
| class | memlog |
| Logging into matrices in data format in memory. More... | |
| class | dirfilelog |
| Logging into dirfile with buffer in memory. More... | |
| class | UIFile |
| class | UIbuilder |
| Builds computational object from a UserInfo structure. More... | |
| class | UIexternal |
| class | UIinternal |
Typedefs | |
| typedef std::map< string, int > | RVmap |
|
typedef map< const string, const UIbuilder * > | UImap |
| Internal structure mapping strings to UIBuilder objects. | |
Enumerations | |
| enum | MixEF_METHOD { EM = 0, QB = 1 } |
| enum | RESAMPLING_METHOD { MULTINOMIAL = 0, STRATIFIED = 1, SYSTEMATIC = 3 } |
| Switch between various resampling methods. | |
Functions | |
| double | egiw_bestbelow (egiw Eg, egiw Eg0, double Egll, ivec &indeces) |
| Return the best structure. | |
| std::ostream & | operator<< (std::ostream &os, const KalmanFull &kf) |
| Array< string > | RV_NAMES (RV_BUFFER_STEP) |
| ivec | RV_SIZES (RV_BUFFER_STEP) |
| std::ostream & | operator<< (std::ostream &os, const RV &rv) |
| RV | concat (const RV &rv1, const RV &rv2) |
| Concat two random variables. | |
| template<class sq_T> | |
| std::ostream & | operator<< (std::ostream &os, mlnorm< sq_T > &ml) |
| UIREGISTER (UIexternal) | |
| UIREGISTER (UIinternal) | |
| void | UI_DBG (Setting &S, const string &spc) |
| [Debugging] Print values in current S to cout | |
| template<class T> | |
| void | UIbuild (Setting &S, T *&ret) |
Prototype of a UI builder. Return value is by the second argument since it type checking via dynamic_cast. | |
| template<class T> | |
| void | UIcall (Setting &S, void(*func)(Setting &, T), T Tmp) |
| Auxiliary function allowing recursivity in S (too complex, remove?). | |
Variables | |
| const int | RV_BUFFER_STEP = 1 |
| RVmap | RV_MAP |
| RV | RV0 = RV() |
| Default empty RV that can be used as default argument. | |
| ivec | RV_SIZES |
| Array< string > | RV_NAMES |
| Uniform_RNG | UniRNG |
| Global Uniform_RNG. | |
| Normal_RNG | NorRNG |
| Global Normal_RNG. | |
| Gamma_RNG | GamRNG |
| Global Gamma_RNG. | |
| UImap | __uimap__ |
| global map of UIbulder names to instances of UIbuilders. Created by UIREGISTER macro | |
![\[ f(rv|rvc) = \frac{f(rv,rvc)}{f(rvc)} \]](form_79.png)
. Function Documentation
| double bdm::egiw_bestbelow | ( | egiw | Eg, | |
| egiw | Eg0, | |||
| double | Egll, | |||
| ivec & | indeces | |||
| ) |
Return the best structure.
- Parameters:
-
Eg a copy of GiW density that is being examined Eg0 a copy of prior GiW density before estimation Egll likelihood of the current Eg indeces current indeces
- Returns:
- best likelihood in the structure below the given one
References ldmat::_D(), ldmat::_L(), bdm::egiw::_V(), ldmat::ldform(), bdm::egiw::lognc(), and ldmat::rows().
Referenced by bdm::ARX::structure_est().
