Mixed-Frequency Bayesian VAR Models

Functions and tools for estimation of mixed-frequency Bayesian vector autoregressive (VAR) models. The package implements a state space-based VAR model that handles mixed frequencies of the data as proposed by Schorfheide and Song (2015) , and extensions thereof developed by Ankargren, Unosson and Yang (2020) , Ankargren and Joneus (2019) , and Ankargren and Joneus (2020) . The models are estimated using Markov Chain Monte Carlo to numerically approximate the posterior distribution. Prior distributions that can be used include normal-inverse Wishart and normal-diffuse priors as well as steady-state priors. Stochastic volatility can be handled by common or factor stochastic volatility models.

Fits Heterogeneous Panel Data Models

Fits heterogeneous panel data models with interactive effects for linear regression, logistic, count, probit, quantile, and clustering. Based on Ando, T. and Bai, J. (2015) "A simple new test for slope homogeneity in panel data models with interactive effects" , Ando, T. and Bai, J. (2015) "Asset Pricing with a General Multifactor Structure" , Ando, T. and Bai, J. (2016) "Panel data models with grouped factor structure under unknown group membership" , Ando, T. and Bai, J. (2017) "Clustering huge number of financial time series: A panel data approach with high-dimensional predictors and factor structures" , Ando, T. and Bai, J. (2020) "Quantile co-movement in financial markets" , Ando, T., Bai, J. and Li, K. (2021) "Bayesian and maximum likelihood analysis of large-scale panel choice models with unobserved heterogeneity" .

Performing Bayesian Inference for Repeated-Measures Designs

A Bayesian credible interval is interpreted with respect to posterior probability, and this interpretation is far more intuitive than that of a frequentist confidence interval. However, standard highest-density intervals can be wide due to between-subjects variability and tends to hide within-subject effects, rendering its relationship with the Bayes factor less clear in within-subject (repeated-measures) designs. This urgent issue can be addressed by using within-subject intervals in within-subject designs, which integrate four methods including the Wei-Nathoo-Masson (2023) , the Loftus-Masson (1994) , the Nathoo-Kilshaw-Masson (2018) , and the Heck (2019) interval estimates.

R Interface to Stan

User-facing R functions are provided to parse, compile, test, estimate, and analyze Stan models by accessing the header-only Stan library provided by the 'StanHeaders' package. The Stan project develops a probabilistic programming language that implements full Bayesian statistical inference via Markov Chain Monte Carlo, rough Bayesian inference via 'variational' approximation, and (optionally penalized) maximum likelihood estimation via optimization. In all three cases, automatic differentiation is used to quickly and accurately evaluate gradients without burdening the user with the need to derive the partial derivatives.

Efficient Leave-One-Out Cross-Validation and WAIC for Bayesian Models

Efficient approximate leave-one-out cross-validation (LOO) for Bayesian models fit using Markov chain Monte Carlo, as described in Vehtari, Gelman, and Gabry (2017) . The approximation uses Pareto smoothed importance sampling (PSIS), a new procedure for regularizing importance weights. As a byproduct of the calculations, we also obtain approximate standard errors for estimated predictive errors and for the comparison of predictive errors between models. The package also provides methods for using stacking and other model weighting techniques to average Bayesian predictive distributions.

Robust Bayesian Meta-Analyses

A framework for estimating ensembles of meta-analytic, meta-regression, and multilevel models (assuming either presence or absence of the effect, heterogeneity, publication bias, and moderators). The RoBMA framework uses Bayesian model-averaging to combine the competing meta-analytic models into a model ensemble, weights the posterior parameter distributions based on posterior model probabilities and uses Bayes factors to test for the presence or absence of the individual components (e.g., effect vs. no effect; Bartoš et al., 2022, ; Maier, Bartoš & Wagenmakers, 2022, ; Bartoš et al., 2025, ). Users can define a wide range of prior distributions for the effect size, heterogeneity, publication bias (including selection models and PET-PEESE), and moderator components. The package provides convenient functions for summary, visualizations, and fit diagnostics.

Computation of Bayes Factors for Common Designs

A suite of functions for computing various Bayes factors for simple designs, including contingency tables, one- and two-sample designs, one-way designs, general ANOVA designs, and linear regression.

OLS, Moderated, Logistic, and Count Regressions Made Simple

Provides SPSS- and SAS-like output for least squares multiple regression, logistic regression, and count variable regressions. Detailed output is also provided for OLS moderated regression, interaction plots, and Johnson-Neyman regions of significance. The output includes standardized coefficients, partial and semi-partial correlations, collinearity diagnostics, plots of residuals, and detailed information about simple slopes for interactions. The output for some functions includes Bayes Factors and, if requested, regression coefficients from Bayesian Markov Chain Monte Carlo analyses. There are numerous options for model plots. The REGIONS_OF_SIGNIFICANCE function also provides Johnson-Neyman regions of significance and plots of interactions for both lm and lme models. There is also a function for partial and semipartial correlations and a function for conducting Cohen's set correlation analyses.

MCMC, Particle Filtering, and Programmable Hierarchical Modeling

A system for writing hierarchical statistical models largely compatible with 'BUGS' and 'JAGS', writing nimbleFunctions to operate models and do basic R-style math, and compiling both models and nimbleFunctions via custom-generated C++. 'NIMBLE' includes default methods for MCMC, Laplace Approximation, Monte Carlo Expectation Maximization, and some other tools. The nimbleFunction system makes it easy to do things like implement new MCMC samplers from R, customize the assignment of samplers to different parts of a model from R, and compile the new samplers automatically via C++ alongside the samplers 'NIMBLE' provides. 'NIMBLE' extends the 'BUGS'/'JAGS' language by making it extensible: New distributions and functions can be added, including as calls to external compiled code. Although most people think of MCMC as the main goal of the 'BUGS'/'JAGS' language for writing models, one can use 'NIMBLE' for writing arbitrary other kinds of model-generic algorithms as well. A full User Manual is available at < https://r-nimble.org>.

Bayesian Regression Models using 'Stan'

Fit Bayesian generalized (non-)linear multivariate multilevel models using 'Stan' for full Bayesian inference. A wide range of distributions and link functions are supported, allowing users to fit -- among others -- linear, robust linear, count data, survival, response times, ordinal, zero-inflated, hurdle, and even self-defined mixture models all in a multilevel context. Further modeling options include both theory-driven and data-driven non-linear terms, auto-correlation structures, censoring and truncation, meta-analytic standard errors, and quite a few more. In addition, all parameters of the response distribution can be predicted in order to perform distributional regression. Prior specifications are flexible and explicitly encourage users to apply prior distributions that actually reflect their prior knowledge. Models can easily be evaluated and compared using several methods assessing posterior or prior predictions. References: Bürkner (2017) ; Bürkner (2018) ; Bürkner (2021) ; Carpenter et al. (2017) .