pf
|
A class template for Gamma filtering. More...
#include <cf_filters.h>
Public Types | |
using | psv = Eigen::Matrix< float_t, dim_pred, 1 > |
"predictor size vector" | |
using | tsv = Eigen::Matrix< float_t, 2, 1 > |
"two by 1 vector" | |
![]() | |
using | obs_sized_vec = Eigen::Matrix< float_t, dimobs, 1 > |
using | state_sized_vec = Eigen::Matrix< float_t, dimstate, 1 > |
Public Member Functions | |
gamFilter (const float_t &nOneTilde, const float_t &dOneTilde) | |
Default constructor. More... | |
virtual | ~gamFilter () |
The (virtual) desuctor. | |
float_t | getLogCondLike () const |
Get the latest conditional likelihood. More... | |
tsv | getFilterVec () const |
Get the current filter vector. More... | |
void | update (const float_t &yt, const psv &xt, const psv &beta, const float_t &sigmaSquared, const float_t &delta) |
Perform a filtering update. More... | |
![]() | |
virtual | ~cf_filter () |
The (virtual) destructor. | |
virtual float_t | getLogCondLike () const=0 |
returns the log of the most recent conditional likelihood More... | |
Private Attributes | |
tsv | m_filtVec |
filter vector (shape and rate) | |
float_t | m_lastLogCondLike |
last log of the conditional likelihood | |
bool | m_fresh |
has data been observed? | |
A class template for Gamma filtering.
pf::filters::gamFilter< dim_pred, float_t >::gamFilter | ( | const float_t & | nOneTilde, |
const float_t & | dOneTilde | ||
) |
Default constructor.
Need ths fir constructing default std::array<>s. Fills all vectors and matrices with zeros. Constructor
nOneTilde | degrees of freedom for time 1 prior. |
dOneTilde | rate parameter for time 1 prior. |
auto pf::filters::gamFilter< dim_pred, float_t >::getFilterVec |
Get the current filter vector.
get the current filtering distribution. First element is the shape, second is the rate.
auto pf::filters::gamFilter< dim_pred, float_t >::getLogCondLike |
Get the latest conditional likelihood.
void pf::filters::gamFilter< dim_pred, float_t >::update | ( | const float_t & | yt, |
const psv & | xt, | ||
const psv & | beta, | ||
const float_t & | sigmaSquared, | ||
const float_t & | delta | ||
) |
Perform a filtering update.
Perform a Gamma filter update.
yt | the most recent dependent random variable |
xt | the most recent predictor vector |
beta | the beta vector |
sigmaSquared | the observation variance scale parameter. |
delta | between 0 and 1 the discount parameter |