Learn R Programming

SKFCPD (version 0.2.4)

Fast Online Changepoint Detection for Temporally Correlated Data

Description

Sequential Kalman filter for scalable online changepoint detection by temporally correlated data. It enables fast single and multiple change points with missing values. See the reference: Hanmo Li, Yuedong Wang, Mengyang Gu (2023), .

Copy Link

Version

Install

install.packages('SKFCPD')

Monthly Downloads

141

Version

0.2.4

License

GPL (>= 3)

Maintainer

Hanmo Li

Last Published

February 17th, 2024

Functions in SKFCPD (0.2.4)

get_LY_online

Updating Kalman filter parameters
SKFCPD

Getting the results of the SKFCPD model
Get_Q_K

matrices and vectors for the inverse covariance in the predictive distribution
compute_log_lik

Natural logarithm of profile likelihood by the fast computing algorithm
productPowerMinusHalf

Square root of product for every elements in a vector
plot_SKFCPD

Plot for SKFCPD model
GaSP_CPD_pred_dist_objective_prior_direct_online

Computing the predictive distribution directly in the online fashion
get_predictive_dist_direct_objective_prior

Updating the predictive distribution
get_predictive_dist_KF_objective_prior

Updating the predictive distribution
get_mu_sigma_hat

Caculate the mean and variance parameter through fast computation
SKFCPD-package

Dynamic Linear Model for Online Changepoint Detection
CPD_DLM

Setting up the CPD_DLM model
Construct_W_matern_5_2_fastGP

The conditional covariance matrix for matern covariance with roughness parameter 2.5
Get_log_det_S2_one_dim

the natural logarithm of the determinant of the correlation matrix and the estimated sum of squares in the exponent of the profile likelihood
GaSP_CPD_pred_dist_objective_prior_KF_online

Computing the predictive distribution in the online fashion
Construct_G_matern_5_2_fastGP

The coefficient matrix in the dynamic linear model when kernel is the Matern covariance with roughness parameter 2.5.
KF_ini

Getting inital Kalman filter parameters
Construct_G_matern_5_2_one_dim

The coefficient matrix in the dynamic linear model when kernel is the Matern covariance with roughness parameter 2.5.
Construct_W0_exp_one_dim

covariance of the stationary distribution of the state when kernel is the exponential covariance.
Estimate_GP_params

Estimate parameters from fast computation of GaSP model
Construct_W_matern_5_2_one_dim

The conditional covariance matrix for matern covariance with roughness parameter 2.5
Construct_W0_matern_5_2_one_dim

covariance of the stationary distribution of the state when kernel is the Matern covariance with roughness parameter 2.5.
Construct_G_exp_fastGP

The coefficient matrix in the dynamic linear model when kernel is the exponential covariance
KF_param_update_for_profile_like

Updating Kalman filter parameters
Construct_W_exp_fastGP

The conditional covariance matrix of the state in the dynamic linear model when kernel is the exponential covariance
KF_ini_for_profile_like

Getting inital Kalman filter parameters for different observation sequences
SKFCPD-class

Class "SKFCPD"
Construct_G_W_W0_V

Generating coefficient and conditional matrics
Estimated_GP_params-class

Estimated GaSP parameters class