KFAS - Kalman Filter and Smoother for Exponential Family State Space Models
State space modelling is an efficient and flexible framework for statistical inference of a broad class of time series and other data. KFAS includes computationally efficient functions for Kalman filtering, smoothing, forecasting, and simulation of multivariate exponential family state space models, with observations from Gaussian, Poisson, binomial, negative binomial, and gamma distributions. See the paper by Helske (2017) <doi:10.18637/jss.v078.i10> for details.
Last updated 5 months ago
dynamic-linear-modelexponential-familyfortrangaussian-modelsstate-spacetime-seriesopenblas
10.49 score 64 stars 16 dependents 242 scripts 5.2k downloadsseqHMM - Mixture Hidden Markov Models for Social Sequence Data and Other Multivariate, Multichannel Categorical Time Series
Designed for fitting hidden (latent) Markov models and mixture hidden Markov models for social sequence data and other categorical time series. Also some more restricted versions of these type of models are available: Markov models, mixture Markov models, and latent class models. The package supports models for one or multiple subjects with one or multiple parallel sequences (channels). External covariates can be added to explain cluster membership in mixture models. The package provides functions for evaluating and comparing models, as well as functions for visualizing of multichannel sequence data and hidden Markov models. Models are estimated using maximum likelihood via the EM algorithm and/or direct numerical maximization with analytical gradients. All main algorithms are written in C++ with support for parallel computation. Documentation is available via several vignettes in this page, and the paper by Helske and Helske (2019, <doi:10.18637/jss.v088.i03>).
Last updated 2 years ago
categorical-dataem-algorithmhidden-markov-modelshmmmixture-markov-modelstime-seriesopenblascppopenmp
8.51 score 97 stars 1 dependents 93 scripts 435 downloadswalker - Bayesian Generalized Linear Models with Time-Varying Coefficients
Efficient Bayesian generalized linear models with time-varying coefficients as in Helske (2022, <doi:10.1016/j.softx.2022.101016>). Gaussian, Poisson, and binomial observations are supported. The Markov chain Monte Carlo (MCMC) computations are done using Hamiltonian Monte Carlo provided by Stan, using a state space representation of the model in order to marginalise over the coefficients for efficient sampling. For non-Gaussian models, the package uses the importance sampling type estimators based on approximate marginal MCMC as in Vihola, Helske, Franks (2020, <doi:10.1111/sjos.12492>).
Last updated 5 months ago
bayesiangeneralized-linear-modelsmcmcstantime-seriesopenblascpp
6.42 score 44 stars 15 scripts 383 downloadsbssm - Bayesian Inference of Non-Linear and Non-Gaussian State Space Models
Efficient methods for Bayesian inference of state space models via Markov chain Monte Carlo (MCMC) based on parallel importance sampling type weighted estimators (Vihola, Helske, and Franks, 2020, <doi:10.1111/sjos.12492>), particle MCMC, and its delayed acceptance version. Gaussian, Poisson, binomial, negative binomial, and Gamma observation densities and basic stochastic volatility models with linear-Gaussian state dynamics, as well as general non-linear Gaussian models and discretised diffusion models are supported. See Helske and Vihola (2021, <doi:10.32614/RJ-2021-103>) for details.
Last updated 5 months ago
bayesian-inferencecppmarkov-chain-monte-carloparticle-filterstate-spacetime-seriesopenblascppopenmp
6.26 score 41 stars 11 scripts 971 downloadsRlibeemd - Ensemble Empirical Mode Decomposition (EEMD) and Its Complete Variant (CEEMDAN)
An R interface for libeemd (Luukko, Helske, Räsänen, 2016) <doi:10.1007/s00180-015-0603-9>, a C library of highly efficient parallelizable functions for performing the ensemble empirical mode decomposition (EEMD), its complete variant (CEEMDAN), the regular empirical mode decomposition (EMD), and bivariate EMD (BEMD). Due to the possible portability issues CRAN version no longer supports OpenMP, you can install OpenMP-supported version from GitHub: <https://github.com/helske/Rlibeemd/>.
Last updated 1 years ago
cdecompositioneemdemdtime-seriesgslcppopenmp
6.11 score 39 stars 13 dependents 17 scripts 502 downloadsramcmc - Robust Adaptive Metropolis Algorithm
Function for adapting the shape of the random walk Metropolis proposal as specified by robust adaptive Metropolis algorithm by Vihola (2012) <doi:10.1007/s11222-011-9269-5>. The package also includes fast functions for rank-one Cholesky update and downdate. These functions can be used directly from R or the corresponding C++ header files can be easily linked to other R packages.
Last updated 3 years ago
openblascpp
6.03 score 6 stars 12 dependents 8 scripts 984 downloadschanger - Change R Package Name
Changing the name of an existing R package is annoying but common task especially in the early stages of package development. This package (mostly) automates this task.
Last updated 3 years ago
4.39 score 49 stars 2 scripts 176 downloadsdiagis - Diagnostic Plot and Multivariate Summary Statistics of Weighted Samples from Importance Sampling
Fast functions for effective sample size, weighted multivariate mean, variance, and quantile computation, and weight diagnostic plot for generic importance sampling type or other probability weighted samples.
Last updated 1 years ago
cppimportance-samplingweighted-samplesopenblascpp
4.32 score 1 stars 1 dependents 14 scripts 643 downloadstsPI - Improved Prediction Intervals for ARIMA Processes and Structural Time Series
Prediction intervals for ARIMA and structural time series models using importance sampling approach with uninformative priors for model parameters, leading to more accurate coverage probabilities in frequentist sense. Instead of sampling the future observations and hidden states of the state space representation of the model, only model parameters are sampled, and the method is based solving the equations corresponding to the conditional coverage probability of the prediction intervals. This makes method relatively fast compared to for example MCMC methods, and standard errors of prediction limits can also be computed straightforwardly.
Last updated 1 years ago
fortranopenblas
3.60 score 8 stars 6 scripts 196 downloadsparticlefield - Sequential Monte Carlo for Latent Conditional Autoregressive Model
Functions for replicating the results of the latent Gaussian Markov random field experiment of Lindsten, Helske, Vihola (2018), XX. Contains also functions for performing particle Markov chain Monte Carlo estimation of the model parameters.
Last updated 4 years ago
cpp
2.18 score 3 stars 4 scripts