FluxPoint0.1.2 package

Change Point Detection for Non-Stationary and Cross-Correlated Time Series

add_jumps

Add mean shifts to multivariate time series data

applyThreshold

Apply Thresholding to VAR Coefficients

computeResiduals

Compute VAR Model Residuals

cvVAR_ENET

Cross Validation for Elastic Net VAR Estimation

cvVAR

Cross-Validated VAR Estimation using Elastic Net

duplicateMatrix

Construct Lagged Design Matrix for VAR

estimate_mus

Estimate the fluctuating mean sequence via maximum likelihood

estimate_musseg

Estimate fluctuating mean segmentwise given detected change points

estimate_RWVAR_cp_heter

Robust parameter estimation (RPE) for multivariate time series

estimate_RWVAR_cp

Robust parameter estimation (RPE) for univariate time series

estimateCovariance

Estimate Covariance Matrix from Residuals

estimatePhinu_nondiag

Estimate non-diagonal VAR(1) parameters after mean removal

fitVAR

Fit VAR Model with Elastic Net via Cross Validation

FluxPoint_raw

Core change point detection algorithm (given known parameters)

FluxPoint

FluxPoint change point detection algorithm

generate_data

Generate multivariate time series from the proposed model

get_metrics

Evaluate change point detection accuracy metrics

get_Sig_e1_approx

Approximate the long-run covariance matrix $\Gamma_{\boldsymbol{\epsil...

get_Sigs

Compute the covariance matrix ΣALL\Sigma_{\mathrm{ALL}}^* for observatio...

inver

Matrix inverse

objective_func

Objective function for robust parameter estimation (RPE)

plot_FluxPoint

Plot multivariate time series with detected change points and estimate...

random_Phi

Randomly generate an autoregressive coefficient matrix Φ\Phi

random_Signu

Randomly generate an innovation covariance matrix $\Sigma_{\boldsymbol...

splitMatrix

Split Coefficient Matrix into VAR Lags

sqrtmat

Matrix square root

transformData

Transform Data for VAR Estimation

Implements methods for multiple change point detection in multivariate time series with non-stationary dynamics and cross-correlations. The methodology is based on a model in which each component has a fluctuating mean represented by a random walk with occasional abrupt shifts, combined with a stationary vector autoregressive structure to capture temporal and cross-sectional dependence. The framework is broadly applicable to correlated multivariate sequences in which large, sudden shifts occur in all or subsets of components and are the primary targets of interest, whereas small, smooth fluctuations are not. Although random walks are used as a modeling device, they provide a flexible approximation for a wide class of slowly varying or locally smooth dynamics, enabling robust performance beyond the strict random walk setting.

  • Maintainer: Yuhan Tian
  • License: GPL-2
  • Last published: 2026-01-10