🌟 The design includes: 💛 suite of models 💛 econometric and numerical methods 💛 algorithms in C++ 💛 handling of inputs and outputs 💛 simple workflows 🌐 doi.org/10.48550/arX... #forecasting #rstats #bsvars #bsvarSIGNs #bvars

⭐ HA! A new working paper on software design for our R packages bsvars, bsvarSIGNs, and bvars is out! ✨ Have a look! 🌐 doi.org/10.48550/arX... #forecasting #rstats #bsvars #bsvarSIGNs #bvars

⭐ Have a look at Adam's presentation of our new working paper on the software design for our R packages bsvars, bsvarSIGNs, and bvars! 💛 It's all about 💛 models 💛 techniques 💛 C++ 💛 IO 💛 workflows 🌐 bsvars.org/2026-08-conc... 🌐 doi.org/10.48550/arX... #forecasting #rstats #bsvars #bsvarSIGNs #bvars

⭐ This holds for ⭐ structural VARs and common volatility BVARs ⭐ point and density forecasts ⭐ 1-month and 1-year ahead forecasts 🌐 doi.org/10.48550/arX... #forecasting #bvars #sv #rstats #bsvars #bsvarSIGNs #bvars

⭐ Alright! In our new working paper we show that ⭐⭐ Bayesian Vector Autoregressions with non-centred Stochastic Volatility forecast better than those with centred Stochastic Volatility! ⭐⭐ ⭐ That's huge! 🌐 doi.org/10.48550/arX... #forecasting #bvars #sv #rstats #bsvars #bsvarSIGNs #bvars

💛🩷 the methods implemented in bsvars version 4.0 are based on sound research 💛🩷 a methodological paper for the package was published in the Journal of Econometrics 💛🩷 see Lütkepohl, Shang, Uzeda, Woźniak (2026) doi.org/10.1016/j.je... #bsvars #rstats #econsky

💛🩷 the bsvars package development gets collaborative! 💛🩷 Fei did code revision 💛🩷 Xiaolei and Bruno sent PRs 💛🩷 Rob suggested improvements 💛🩷 Helmut, Luis and Fei coauthored a paper 💛🩷 cran.r-project.org/package=bsvars @adamwang15.bsky.social @robjhyndman.com #bsvars #rstats #cooperation #econsky

💛🩷 install.packages("bsvars") #bsvars #rstats

🩷💛 a new exogeneous volatility regimes model available in bsvars 4.0 🩷💛 choose when volatility changes 🩷💛 commonly used 🩷💛 might work well with data 🩷💛 fast estimation 🩷💛 see Brunnermeier, Palia, Sasry, Sims (2021, AER) 🩷💛 see Shang, Woźniak (2026) arxiv.org/abs/2603.16035 #bsvars #rstats #econsky

💛🩷 a new feature of Student-t shocks available in bsvars 4.0 💛🩷 all models may feature Student-t shocks 💛🩷 source of identification 💛🩷 verify identification 💛🩷 improved forecasting 💛🩷 ensured normalisation 💛🩷 see Shang, Wang, Woźniak (2026) - coming up soon #bsvars #rstats #econsky

💛🩷 see Shang, Woźniak (2026) doi.org/10.48550/arX...

💛🩷 identification verification - a new feature bsvars 4.0 💛🩷 verify identification via heteroskedasticity 💛🩷 verify non-normality 💛🩷 for all volatility models 💛🩷 flexible specification 💛🩷 see Lütkepohl, Shang, Uzeda, Woźniak (2026) doi.org/10.1016/j.je... #bsvars #rstats #econsky

🩷🖤 Access all the features of the new version of the R package bsvarSIGNs for Bayesian Estimation of Structural Vector Autoregressions Identified by Sign, Zero, and Narrative Restrictions by typing: 🩷🖤 install.packages("bsvarSIGNs") 🩷🖤 cran.r-project.org/package=bsva... #bsvarSIGNs #rstats #econsky

🩷💛 new features in version 4.0 of the R package bsvars 🩷💛 new volatility models 🩷💛 Student-t shocks 🩷💛 identification verification 🩷💛 improved plot and summary 🩷💛 code improvements 🩷💛 bsvars.org/bsvars/ #bsvars #rstats #foss #econsky

See more in our recent working paper: arxiv.org/abs/2603.16035

Identification Verification for Structural Vector Autoregressions with Sparse Heterogeneous Markov Switching Heteroskedasticity

We propose a structural vector autoregressive model with a new and flexible specification of the volatility process which we call Sparse Heterogeneous Markov-Switching Heteroskedasticity. In this mode...

arxiv.org

🩷💛 a new Markov switching volatility model available in bsvars 4.0 🩷💛 heterogeneous specification 🩷💛 a Markov process per shock 🩷💛 sparse specification 🩷💛 excellent forecasting 🩷💛 precise estimation 🩷💛 see Shang, Woźniak (2026) 🩷💛 arxiv.org/abs/2603.16035 #bsvars #rstats #econsky

Updates on CRAN: actinet (0.4.0), bsvars (4.0), colorr (1.1.0), dann (1.2.0), eiIT (0.0.2-1), funcml (0.9.0), glyrepr (1.0.0), gorica (0.1.6), GPArotation (2026.8-2), GRIN2 (2.1.0), heims (0.4.3), hexify (0.8.2)

🩷💛 A new version 4.0 of the R package bsvars has just landed on CRAN! 🩷💛 Bayesian Estimation of Structural Vector Autoregressive Models 🩷💛 With plenty new features and models! 🩷💛 It's awesome! 🩷💛 cran.r-project.org/package=bsvars #bsvars #rstats #foss #econsky

🩷💛 A new version 4.0 of the R package bsvars has just landed on CRAN! 🩷💛 Bayesian Estimation of Structural Vector Autoregressive Models 🩷💛 With plenty new features and models! 🩷💛 It's awesome! 🩷💛 bsvars.org/bsvars/ #bsvars #rstats #foss

💜🖤 bsvarSIGNs new feature: COVID-specific volatility 💜🖤 The package uses a flexible BVAR with estimated hyper-parameters of the Minnesota prior by Giannone, Lenza & Primiceri (2015, RESTAT) 💜🖤 The model is NOW extended by COVID-specific volatility as in Lenza & Primiceri (2022, JAE) 💜🖤 bsvars.org

💜🖤 bsvarSIGNs new feature: small-open economy 💜🖤 The package NOW features a small-open economy model 💜🖤 It is estimated with block-exogeneity restrictions on structural and autoregressive parameters 💜🖤 It allows sign and narrative restrictions 💜🖤 bsvars.org/bsvarSIGNs/ #bsvarSIGNs #rstats

💜🖤 bsvarSIGNs new feature: parallel computations 💜🖤 We implemented parallel computations for posterior sampling 💜🖤 This is possible thanks to the particular model specification 💜🖤 Model estimation is even faster NOW 💜🖤 bsvars.org/bsvarSIGNs/ #bsvarSIGNs #rstats

Bayesian SVARs with Sign, Zero, and Narrative Restrictions

Implements state-of-the-art algorithms for the Bayesian analysis of Structural Vector Autoregressions (SVARs) identified by sign, zero, and narrative restrictions. The core model is based on a flexibl...

bsvars.org

🩷🖤 NEW FEATURES in version 3.0 of bsvarSIGNs 🩷 BVAR with COVID-specific volatility as in Lenza, Primiceri (2022, JAE) 🖤 Small-open economy model with sign and narrative restrictions 🩷 Even faster parallel computations for posterior sampling 🖤 More info at: bsvars.org/bsvarSIGNs/ #bsvarSIGNs #rstats

🖤🩷 WOW! 🩷🖤 🖤🩷 Version 3.0 of our R package bsvarSIGNs is on CRAN now! 🩷🖤 🖤🩷 And it has new superb features! 🩷🖤 🌐 have a look: bsvars.org/bsvarSIGNs/ #bsvarSIGNs #rstats #foss #structural #BVARs #macro #forecasting #econsky

🖤🩷 Big news! bsvarSIGNs version 3.0 is on CRAN now! And it's great! 🩷🖤 🌐 have a look: cran.r-project.org/package=bsva... Thanks @adamwang15.bsky.social #bsvarSIGNs #rstats #foss #structural #BVARs #macro #forecasting #econsky

⭐⭐⭐ bsvars has reached 30K downloads! ⭐⭐⭐ ⭐⭐⭐ Infinite thanks to all the users! Thank you! ⭐⭐⭐ ⭐⭐⭐ bsvars.org ⭐⭐⭐ ⭐ Watch this space! There's lots of new things coming! #bsvars #rstats #foss #econsky

🌟🌟🌟 bsvarSIGNs has reached 10K downloads 🌟🌟🌟 🌟🌟🌟 Thanks to all of you who downloaded and used the package! 🌟🌟🌟 🌟🌟🌟 bsvars.org 🌟🌟🌟 🌟 More goodies is coming soon! #bsvarSIGNs #rstats #foss #econsky

💝 Ah, it's so good when co-authors go through the final code review! 🛍️ Thanks Fei for correcting the forecasts! 💖 github.com/bsvars/bsvar... #bsvars #forecasting #cooperation #foss #econsky

Fix recursive forecasts for non-centred SV and Student-t specs by lcq110 · Pull Request #143 · bsvars/bsvars

I found two issues in src/forecast.cpp. 1. Non-centred SV forecasts The current code has: ht(n) = posterior_omega(n, s) * (posterior_rho(n, s) * ht(n) + xx); forecasts_sigma2(n, h, s) = exp...

github.com

🖤💙 Save the date! 💙💛 My Workshop for Ukraine organised by @dariia.bsky.social is "All About bpvars" and takes place on Sep 10! Pls remember to register and see you then! 🌐 sites.google.com/view/dariia-... #w4UKR #bpvars #rstats

Dariia Mykhailyshyna - Workshops for Ukraine

Feedback on the past workshops (if you want to learn how to make wordclouds, check out Text Data Analysis workshop below)

sites.google.com

🖤💙 Save the date! 💙💛 My Workshop for Ukraine organised by @dariia.bsky.social is "All About bpvars" and takes place on Sep 10! Pls remember to register and see you then! 🌐 sites.google.com/view/dariia-... #w4UKR #bpvars #rstats