Changelog
Source:NEWS.md
bvhar 2.0.1
Fix internal vectorization and unvectorization behavior.
Used Eigen 3.4 feature (
reshaped()
) to solve these (RcppEigen >= 0.3.4.0.0
).
bvhar 2.0.0
CRAN release: 2024-02-14
Start to implement OOP in C++ source for each model, ready for major update.
Add SV specification (
sv_spec
argument) inbvhar_sv()
andbvar_sv()
(set_sv()
).Prevent SSVS overflow issues by using log-sum-exp trick when computing Bernoulli posterior probability.
Add separate constant term prior specification (
intercept
) inbvhar_sv()
andbvar_sv()
(set_intercept()
).Convert every header file inst/include to header-only format. This enables external inclusion of our classes, structs, and Rcpp functions by using
LinkingTo
(in R package development) or// [[Rcpp::depends(RcppEigen, BH, bvhar)]]
.
Parallel Chain MCMC
Use OpenMP parallel for loop
Progress bar will show the status only for master thread when OpenMP enabled.
Interruption detect will just save values and break the loop, not return immediately.
Do burn-in and thinning in each
returnRecords()
method to make pre-process parallel chains easier.Use boost library (
BH
package) RNG instead of Rf_* RNG ofRcpp
for thread-safety.Introduce function overloading to internal Rcpp random generation functions temporarily. It’s for maintaining
set.seed()
usage of some functions.
bvhar 1.2.0
CRAN release: 2024-01-09
Replace progress bar of
RcppProgress
package with custom header (bvharprogress.h
).Replace checking user interruption in the same package with custom header (
bvharinterrupt.h
).Fix triangular algorithm. Found missing update of some variables (
bvar_sv()
andbvhar_sv()
).
bvhar 1.1.0
CRAN release: 2023-12-18
For new research, add new features for shrinkage priors.
Add Shrinkage priors SSVS and Horseshoe (
bvar_ssvs()
,bvhar_ssvs()
,bvar_horseshoe()
, andbvhar_horseshoe()
).bvar_sv()
,bvhar_sv()
works with SSVS (set_ssvs()
) and Horseshoe (set_horseshoe()
).Update the shrinkage structure in the spirit of Minnesota. (
minnesota = TRUE
,minnesota = c("no", "short", "longrun")
).Stochastic volatility models implement corrected triangular algorithm of Carriero et al. (2021).
bvhar 1.0.2
CRAN release: 2023-12-06
License has been changed to GPLv3.
Remove unnecessary Rcpp plugins in source files.
bvhar 1.0.0
CRAN release: 2023-11-08
“Bayesian Vector Heterogeneous Autoregressive Modeling” has been accepted in JSCS 🎉
Update to major version before publication.
bvhar 0.14.0
Add Stochastic Search Variable Selection (SSVS) models for VAR and VHAR (
bvar_ssvs()
andbvhar_ssvs()
)Can do corresponding variable selection (
summary.ssvsmod()
)
bvhar 0.13.0
- Add stochastic volatility models VAR-SV and VHAR-SV (
bvar_sv()
andbvhar_sv()
).
bvhar 0.12.1
Fix not working Hierarchical natural conjugate MNIW function (
bvar_niwhm()
).Use
posterior
package forsummary.normaliw()
to improve processing and printing.
bvhar 0.12.0
Now can use heavy-tailed distribution (Multivariate t-distribution) when generating VAR and VHAR process (
sim_var()
andsim_vhar()
).Also provide independent MVT generation function (
sim_mvt()
).
bvhar 0.10.0
Add partial t-test for each VAR and VHAR coefficient (
summary.varlse()
andsummary.vharlse()
).Appropriate print method for the updated summary method (
print.summary.varlse()
andprint.summary.vharlse()
).
bvhar 0.9.0
Can compute impulse response function for VAR (
varlse
) and VHAR (vharlse
) models (analyze_ir()
).Can draw impulse -> response plot in grid panels (
autoplot.bvharirf()
).
bvhar 0.8.0
Changed the way of specifying the lower and upper bounds of empirical bayes (
bound_bvhar()
).Added Empirical Bayes vignette.
bvhar 0.7.0
- Add one integrated function that can do empirical bayes (
choose_bayes()
andbound_bvhar()
).
bvhar 0.6.0
Added weekly and monthly order feature in VHAR family (
vhar_lm()
andbvhar_minnesota()
).Other functions are compatible with har order option (
predict.vharlse()
,predict.bvharmn()
, andchoose_bvhar()
)
bvhar 0.5.2
- Added parallel option for empirical bayes (
choose_bvar()
andchoose_bvhar()
).
bvhar 0.5.1
- Added facet feature for the loss plot and changed its name (
gg_loss()
).
bvhar 0.5.0
Added rolling window and expanding window features (
forecast_roll()
andforecast_expand()
).Can compute loss for each rolling and expanding window method (
mse.bvharcv()
,mae.bvharcv()
,mape.bvharcv()
, andmape.bvharcv()
).
bvhar 0.4.1
Fix Marginal likelihood form (
compute_logml()
).Optimize empirical bayes method using stabilized marginal likelihood function (
logml_stable()
).
bvhar 0.4.0
Change the way to compute the CI of BVAR and BVHAR (
predict.bvarmn()
,predict.bvharmn()
, andpredict.bvarflat()
)Used custom random generation function - MN, IW, and MNIW based on RcppEigen
bvhar 0.3.0
Added Bayesian model specification functions and class (
bvharspec
).Replaced hyperparameters with model specification in Bayesian models (
bvar_minnesota()
,bvar_flat()
, andbvhar_minnesota()
).
bvhar 0.2.0
- Added constant term choice in each function (
var_lm()
,vhar_lm()
,bvar_minnesota()
,bvar_flat()
, andbvhar_minnesota()
).