Movatterモバイル変換

bsvars 3.2

The package is under intensive development, and more functionalitywill be provided soon! To see the packageROADMAP towardsthe next version.

Have a question, or suggestion, or wanna get in touch? Join thepackageDISCUSSIONforum.

1. The package includes the first version of the vignette5
2. Updates on the websitehttps://bsvars.org/bsvars/
3. New plots with axes reading variable names, time scale, and lettingyou specify structural shock names!97
4. Improved examples for forecasting with exogenous variables. Samplematrices included in the package. Fixed the bug incppcode for forecasting. Thanks to@DawievLill for asking for clarity!96

bsvars 3.1

1. A NEW MODEL! An SVAR with t-distributed structural shocksfacilitating identification through non-normality is now included in thepackage with all the necessary functionality#84
2. New ways of verifying identification through heteroskedasticity ornon-normality using methodverify_identification()#84
3. Improve coding offorecastcppfunction andR methods#89
4. Included or updated legend in FEVD and HD plots as requested by@ccoleman9#85

bsvars 3.0.1

1. Fixed the bugs that started coming up in the new tested version ofArmadillo andRcppArmadillo#82 andRcppCore/RcppArmadillo#443
2. Corrected the computations ofverify_autoregression#82

bsvars 3.0

1. The package has a logo! And it’s beautiful!#37
2. The package includessummary methods#1
3. The package includesplot methods#36
4. Methodforecast allow for conditional forecasting givenprovided future trajectories of selected variables#76
5. Sparse mixture and Markov-switching models can now have more than 20regimes#57
6. A new, more detailed, package description#62
7. The website features the new logo. And includes some new information#38
8. Updates on documentation to accommodate the fact that some genericsand functions from packagebsvars will be used in abroader family of packages, first of which isbsvarSIGNs.Includes updates on references.#63
9. Fixedcompute_fitted_values(). Now it’s correctlysampling from the predictive data density.#67
10. Fixed some bugs that did not create problems#55
11. Got rid of filling by reference in the samplers for the sake ofgranting the exportedcpp functions usability#56
12. Codedcompute_*() functions as generics and methods#70
13. Updated code for forecast error variance decompositions forheteroskedastic models (qas prompted by@adamwang15)#69

bsvars 2.1.0

Published on 11 December 2023

1. Included Bayesian procedure for verifying structural shocks’heteroskedasticty equation-by-equation using Savage-Dickey densityratios#26
2. Included Bayesian procedure for verifying joint hypotheses onautoregressive parameters using Savage-Dickey density ratios#26
3. Included the possibility of specifying exogenous variables ordeterministic terms and included the deterministic terms used byLütkepohl, Shang, Uzeda, Woźniak (2023)#45
4. Updated the data as in Lütkepohl, Shang, Uzeda, Woźniak (2023)#45
5. Fixing the compilation problems reportedHERE#48
6. The package has its pkgdown website atbsvars.org/bsvars/#38

bsvars 2.0.0

Published on 23 October 2023

1. Included Imports from packagestochvol
2. Posterior computations for:

• impulse responses and forecast error variance decomposition#3,
• structural shocks and historical decompositions#14
• fitted values#17
• conditional standard deviations#16
• regime probabilities for MS and MIX models#18

3. Implemented faster samplers based on random number generators fromarmadillo viaRcppArmadillo#7
4. Theestimate_bsvar* functions now also normalise theoutput w.r.t. to a structural matrix with positive elements on the maindiagonal#9
5. Changed the order of arguments in theestimate_bsvar*functions withposterior first to facilitate workflowsusing the pipe|>#10
6. Include citation info for the package#12
7. Corrected sampler for AR parameter of the SV equations#19
8. Added samplers from joint predictive densities#15
9. A new centred Stochastic Volatility heteroskedastic process isimplemented#22
10. Introduced a three-level local-global equation-specific priorshrinkage hierarchy for the parameters of matrices and#34
11. Improved checks for correct specification of argumentsS andthin of theestimate methodas enquired by@mfaragd#33
12. Improved the ordinal numerals presentation for thinning in theprogress bar#27

bsvars 1.0.0

Published on 1 September 2022

1. repo transferred from GitLab to GitHub
2. repository is made public
3. version to be premiered on CRAN

bsvars 0.0.2.9000

1. Added a new progress bar for theestimate_bsvar*functions
2. DevelopedR6 classes for model specification andposterior outcomes; model specification includes sub-classes for priors,identifying restrictions, data matrices, and starting values
3. Added a complete package documentation
4. Written help files
5. Developed tests for MCMC reproducibility
6. Included sample data

bsvars 0.0.1.9000

1. cpp scripts are imported, compile, and give noErrors, Warnings, or Notes
2. R wrappers for the functions are fullyoperating
3. full documentation describing package and functions’ functionality[sic!]