Daniele Durante
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Daniele Durante has a total of 44 co-authors. Here are the top co-authors by number of shared publications:
- David Brian Dunson
Department of Statistical Science, Duke University
9 shared publications - Tommaso Rigon
University of Milano-Bicocca
7 shared publications David B. Dunson
6 shared publications- Antonio Canale
Department of Statistics, University of Padova, Padova, Italy
4 shared publications - Sirio Legramanti
Department Decision Sciences and Institute for Data Science and Analytics, Bocconi University
4 shared publications - Augusto Fasano
Collegio Carlo Alberto
3 shared publications
Latest Publications
- 27 October 2026
- Published by Oxford University Press (OUP) in Journal of the Royal Statistical Society Series A: Statistics in Society
Bayesian local clustering of age-period mortality surfaces across multiple countries
Although classical mortality modelling literature has focused on single countries in isolation, recent contributions have progressively moved toward joint statistical models for multiple countries. Besides favouring borrowing of information to improve age-period forecasts, this perspective has also potential to infer local similarities among countries’ mortality patterns in specific age classes and periods, which could unveil unexplored demographic trends, while guiding the design of targeted policies. Advancements along this direction are undermined by the lack of multi-country models capable of incorporating the core structures of age-period mortality surfaces together with clustering patterns among countries that are not global, but rather vary locally across the different combinations of ages and periods. We cover this gap by developing a Bayesian model for log-mortality rates that characterizes the age structure of mortality through a B-spline expansion whose country-specific dynamic coefficients encode both changes of this age structure across periods and also local clustering patterns among countries under a time-dependent random partition prior for these country-specific dynamic coefficients. While flexible, this formulation facilitates tractable posterior inference leveraging a suitably designed Gibbs sampler. The application to mortality data from 14 countries unveils local similarities which highlight both previously recognized demographic phenomena and also yet unexplored patterns.
- 27 October 2026
- Published by Oxford University Press (OUP) in Journal of the Royal Statistical Society Series A: Statistics in Society
Bayesian local clustering of age-period mortality surfaces across multiple countries
Although classical mortality modelling literature has focused on single countries in isolation, recent contributions have progressively moved toward joint statistical models for multiple countries. Besides favouring borrowing of information to improve age-period forecasts, this perspective has also potential to infer local similarities among countries’ mortality patterns in specific age classes and periods, which could unveil unexplored demographic trends, while guiding the design of targeted policies. Advancements along this direction are undermined by the lack of multi-country models capable of incorporating the core structures of age-period mortality surfaces together with clustering patterns among countries that are not global, but rather vary locally across the different combinations of ages and periods. We cover this gap by developing a Bayesian model for log-mortality rates that characterizes the age structure of mortality through a B-spline expansion whose country-specific dynamic coefficients encode both changes of this age structure across periods and also local clustering patterns among countries under a time-dependent random partition prior for these country-specific dynamic coefficients. While flexible, this formulation facilitates tractable posterior inference leveraging a suitably designed Gibbs sampler. The application to mortality data from 14 countries unveils local similarities which highlight both previously recognized demographic phenomena and also yet unexplored patterns.
- 26 March 2026
- Published by Oxford University Press (OUP) in Journal of the Royal Statistical Society Series C: Applied Statistics
- Vol. 75 (5) , 1240-1262
Low-rank bilinear autoregressive models for three-way criminal activity tensors
Criminal activity data are typically available via a three-way tensor encoding the reported frequencies of different crime categories across time and space. The challenges that arise in the design of interpretable, yet realistic, model-based representations of the complex dependencies within and across these three dimensions have led to an increasing adoption of black-box predictive strategies. While this perspective has proved successful in producing accurate forecasts guiding targeted interventions, the lack of interpretable model-based characterizations of the dependence structures underlying criminal activity tensors prevents from inferring the cascading effects of these interventions across the different dimensions. We address this gap through the design of a low-rank bilinear autoregressive model which achieves comparable predictive performance to black-box strategies, while allowing interpretable inference on the dependence structures of reported criminal activities across crime categories, time and space. This representation incorporates the time dimension via an autoregressive construction that accounts for spatial effects and dependencies among crime categories through a separable low-rank bilinear formulation. When applied to Chicago police reports, the model yields remarkable predictive performance and reveals interpretable dependence structures unveiling fundamental crime dynamics. These results facilitate the design of more refined intervention policies informed by the cascading effects of the policy itself.
- 26 March 2026
- Published by Oxford University Press (OUP) in Journal of the Royal Statistical Society Series C: Applied Statistics
- Vol. 75 (5) , 1240-1262
Low-rank bilinear autoregressive models for three-way criminal activity tensors
Criminal activity data are typically available via a three-way tensor encoding the reported frequencies of different crime categories across time and space. The challenges that arise in the design of interpretable, yet realistic, model-based representations of the complex dependencies within and across these three dimensions have led to an increasing adoption of black-box predictive strategies. While this perspective has proved successful in producing accurate forecasts guiding targeted interventions, the lack of interpretable model-based characterizations of the dependence structures underlying criminal activity tensors prevents from inferring the cascading effects of these interventions across the different dimensions. We address this gap through the design of a low-rank bilinear autoregressive model which achieves comparable predictive performance to black-box strategies, while allowing interpretable inference on the dependence structures of reported criminal activities across crime categories, time and space. This representation incorporates the time dimension via an autoregressive construction that accounts for spatial effects and dependencies among crime categories through a separable low-rank bilinear formulation. When applied to Chicago police reports, the model yields remarkable predictive performance and reveals interpretable dependence structures unveiling fundamental crime dynamics. These results facilitate the design of more refined intervention policies informed by the cascading effects of the policy itself.
- Open Access
- 14 January 2026
- Published by Oxford University Press (OUP) in Journal of the Royal Statistical Society Series B: Statistical Methodology
- Vol. 88 (4) , 1181-1209
Skew-symmetric approximations of posterior distributions
Popular deterministic approximations of posterior distributions from, e.g. the Laplace method, variational Bayes and expectation-propagation, generally rely on symmetric families, often taken to be Gaussian. This choice facilitates optimization and inference, but typically affects the quality of the approximation. In fact, even in basic parametric models, posterior distributions often display asymmetries that yield bias and reduced accuracy when considering symmetric approximations. Recent research has moved towards more flexible approximations incorporating skewness. However, current solutions are often model specific, lack general supporting theory, increase the computational complexity of the optimization problem, and do not provide broadly applicable solutions to incorporate skewness in any symmetric approximation. This article addresses such a gap by introducing a general and provably optimal strategy to perturb any symmetric approximation of a generic posterior distribution. The proposed solution is derived without additional optimization steps, and yields a similarly tractable approximation within the class of skew-symmetric densities that enhances the finite sample accuracy of the original symmetric counterpart. Furthermore, under suitable assumptions, it improves the convergence rate to the exact posterior by at least a n factor, in asymptotic regimes. These advancements are illustrated in numerical studies focusing on skewed perturbations of state-of-the-art Gaussian approximations.
- 14 January 2026
- Published by Oxford University Press (OUP) in Journal of the Royal Statistical Society Series B: Statistical Methodology
- Vol. 88 (4) , 1181-1209
Open AccessSkew-symmetric approximations of posterior distributions
Popular deterministic approximations of posterior distributions from, e.g. the Laplace method, variational Bayes and expectation-propagation, generally rely on symmetric families, often taken to be Gaussian. This choice facilitates optimization and inference, but typically affects the quality of the approximation. In fact, even in basic parametric models, posterior distributions often display asymmetries that yield bias and reduced accuracy when considering symmetric approximations. Recent research has moved towards more flexible approximations incorporating skewness. However, current solutions are often model specific, lack general supporting theory, increase the computational complexity of the optimization problem, and do not provide broadly applicable solutions to incorporate skewness in any symmetric approximation. This article addresses such a gap by introducing a general and provably optimal strategy to perturb any symmetric approximation of a generic posterior distribution. The proposed solution is derived without additional optimization steps, and yields a similarly tractable approximation within the class of skew-symmetric densities that enhances the finite sample accuracy of the original symmetric counterpart. Furthermore, under suitable assumptions, it improves the convergence rate to the exact posterior by at least a n factor, in asymptotic regimes. These advancements are illustrated in numerical studies focusing on skewed perturbations of state-of-the-art Gaussian approximations.
- Open Access
- 19 March 2025
- Published by Oxford University Press (OUP) in Journal of the Royal Statistical Society Series A: Statistics in Society
- Vol. 189 (2) , 869-897
Zero-inflated stochastic block modelling of efficiency-security trade-offs in weighted criminal networks
Criminal networks arise from the attempt to balance a need of establishing frequent ties among affiliates to facilitate coordination of illegal activities, with the necessity to sparsify the overall connectivity architecture to hide from law enforcement. This efficiency-security trade-off is also combined with the creation of groups of redundant criminals that exhibit similar connectivity patterns, thus guaranteeing resilient network architectures. State-of-the-art models for such data are not designed to infer these unique structures. In contrast to such solutions, we develop a tractable Bayesian zero-inflated Poisson stochastic block model (ZIP–SBM), which identifies groups of redundant criminals having similar connectivity patterns, and infers both overt and covert block interactions within and across these groups. This is accomplished by modelling the weighted ties (corresponding to counts of interactions among pairs of criminals) via zero-inflated Poisson distributions with block-specific parameters that quantify complex patterns in the excess of zero ties in each block (security) relative to the distribution of the observed weighted ties within that block (efficiency). The performance of ZIP–SBM is illustrated in simulations and in a study of summit co-attendances in a complex Mafia organization, where we unveil efficiency-security structures adopted by the criminal organization that were hidden to previous analyses.
- 19 March 2025
- Published by Oxford University Press (OUP) in Journal of the Royal Statistical Society Series A: Statistics in Society
- Vol. 189 (2) , 869-897
Open AccessZero-inflated stochastic block modelling of efficiency-security trade-offs in weighted criminal networks
Criminal networks arise from the attempt to balance a need of establishing frequent ties among affiliates to facilitate coordination of illegal activities, with the necessity to sparsify the overall connectivity architecture to hide from law enforcement. This efficiency-security trade-off is also combined with the creation of groups of redundant criminals that exhibit similar connectivity patterns, thus guaranteeing resilient network architectures. State-of-the-art models for such data are not designed to infer these unique structures. In contrast to such solutions, we develop a tractable Bayesian zero-inflated Poisson stochastic block model (ZIP–SBM), which identifies groups of redundant criminals having similar connectivity patterns, and infers both overt and covert block interactions within and across these groups. This is accomplished by modelling the weighted ties (corresponding to counts of interactions among pairs of criminals) via zero-inflated Poisson distributions with block-specific parameters that quantify complex patterns in the excess of zero ties in each block (security) relative to the distribution of the observed weighted ties within that block (efficiency). The performance of ZIP–SBM is illustrated in simulations and in a study of summit co-attendances in a complex Mafia organization, where we unveil efficiency-security structures adopted by the criminal organization that were hidden to previous analyses.
- 9 March 2025
- Published by Springer Nature
- p. 3-9
Conjugacy and Approximation Properties of Skew–Symmetric Distributions in Statistical Inference
- 9 March 2025
- Published by Springer Nature
- p. 3-9
Conjugacy and Approximation Properties of Skew–Symmetric Distributions in Statistical Inference
- Open Access
- 1 December 2024
- Published by Institute of Mathematical Statistics in The Annals of Statistics
- Vol. 52 (6) , 2714-2737
Skewed Bernstein–von Mises theorem and skew-modal approximations
Gaussian deterministic approximations are routinely employed in Bayesian statistics to ease inference when the target posterior is intractable. While these approximations are justified, in asymptotic regimes, by Bernstein–von Mises type results, in practice the expected Gaussian behavior might poorly represent the actual shape of the target posterior, thus affecting approximation accuracy. Motivated by these considerations, we derive an improved class of closed-form and valid deterministic approximations of posterior distributions that arise from a novel treatment of a third-order version of the Laplace method yielding approximations within a tractable family of skew-symmetric distributions. Under general assumptions accounting for misspecified models and non-i.i.d. settings, such a family of approximations is shown to have a total variation distance from the target posterior whose convergence rate improves by at least one order of magnitude the one achieved by the Gaussian from the classical Bernstein–von Mises theorem. Specializing this result to the case of regular parametric models shows that the same accuracy improvement can be also established for the posterior expectation of polynomially bounded functions. Unlike available higher-order approximations based on, for example, Edgeworth expansions, our results prove that it is possible to derive closed-form and valid densities which provide a more accurate, yet similarly tractable, alternative to Gaussian approximations of the target posterior, while inheriting its limiting frequentist properties. We strengthen these arguments by developing a practical skew-modal approximation for both joint and marginal posteriors which preserves the guarantees of its theoretical counterpart by replacing the unknown model parameters with the corresponding maximum a posteriori estimate. Simulation studies and real-data applications confirm that our theoretical results closely match the empirical gains observed in practice.
- 1 December 2024
- Published by Institute of Mathematical Statistics in The Annals of Statistics
- Vol. 52 (6) , 2714-2737
Open AccessSkewed Bernstein–von Mises theorem and skew-modal approximations
Gaussian deterministic approximations are routinely employed in Bayesian statistics to ease inference when the target posterior is intractable. While these approximations are justified, in asymptotic regimes, by Bernstein–von Mises type results, in practice the expected Gaussian behavior might poorly represent the actual shape of the target posterior, thus affecting approximation accuracy. Motivated by these considerations, we derive an improved class of closed-form and valid deterministic approximations of posterior distributions that arise from a novel treatment of a third-order version of the Laplace method yielding approximations within a tractable family of skew-symmetric distributions. Under general assumptions accounting for misspecified models and non-i.i.d. settings, such a family of approximations is shown to have a total variation distance from the target posterior whose convergence rate improves by at least one order of magnitude the one achieved by the Gaussian from the classical Bernstein–von Mises theorem. Specializing this result to the case of regular parametric models shows that the same accuracy improvement can be also established for the posterior expectation of polynomially bounded functions. Unlike available higher-order approximations based on, for example, Edgeworth expansions, our results prove that it is possible to derive closed-form and valid densities which provide a more accurate, yet similarly tractable, alternative to Gaussian approximations of the target posterior, while inheriting its limiting frequentist properties. We strengthen these arguments by developing a practical skew-modal approximation for both joint and marginal posteriors which preserves the guarantees of its theoretical counterpart by replacing the unknown model parameters with the corresponding maximum a posteriori estimate. Simulation studies and real-data applications confirm that our theoretical results closely match the empirical gains observed in practice.
- Open Access
Skewed Bernstein-von Mises theorem and skew-modal approximations
Gaussian approximations are routinely employed in Bayesian statistics to ease inference when the target posterior is intractable. Although these approximations are asymptotically justified by Bernstein-von Mises type results, in practice the expected Gaussian behavior may poorly represent the shape of the posterior, thus affecting approximation accuracy. Motivated by these considerations, we derive an improved class of closed-form approximations of posterior distributions which arise from a new treatment of a third-order version of the Laplace method yielding approximations in a tractable family of skew-symmetric distributions. Under general assumptions which account for misspecified models and non-i.i.d. settings, this family of approximations is shown to have a total variation distance from the target posterior whose rate of convergence improves by at least one order of magnitude the one established by the classical Bernstein-von Mises theorem. Specializing this result to the case of regular parametric models shows that the same improvement in approximation accuracy can be also derived for polynomially bounded posterior functionals. Unlike other higher-order approximations, our results prove that it is possible to derive closed-form and valid densities which are expected to provide, in practice, a more accurate, yet similarly-tractable, alternative to Gaussian approximations of the target posterior, while inheriting its limiting frequentist properties. We strengthen such arguments by developing a practical skew-modal approximation for both joint and marginal posteriors that achieves the same theoretical guarantees of its theoretical counterpart by replacing the unknown model parameters with the corresponding MAP estimate. Empirical studies confirm that our theoretical results closely match the remarkable performance observed in practice, even in finite, possibly small, sample regimes.
Open AccessSkewed Bernstein-von Mises theorem and skew-modal approximations
Gaussian approximations are routinely employed in Bayesian statistics to ease inference when the target posterior is intractable. Although these approximations are asymptotically justified by Bernstein-von Mises type results, in practice the expected Gaussian behavior may poorly represent the shape of the posterior, thus affecting approximation accuracy. Motivated by these considerations, we derive an improved class of closed-form approximations of posterior distributions which arise from a new treatment of a third-order version of the Laplace method yielding approximations in a tractable family of skew-symmetric distributions. Under general assumptions which account for misspecified models and non-i.i.d. settings, this family of approximations is shown to have a total variation distance from the target posterior whose rate of convergence improves by at least one order of magnitude the one established by the classical Bernstein-von Mises theorem. Specializing this result to the case of regular parametric models shows that the same improvement in approximation accuracy can be also derived for polynomially bounded posterior functionals. Unlike other higher-order approximations, our results prove that it is possible to derive closed-form and valid densities which are expected to provide, in practice, a more accurate, yet similarly-tractable, alternative to Gaussian approximations of the target posterior, while inheriting its limiting frequentist properties. We strengthen such arguments by developing a practical skew-modal approximation for both joint and marginal posteriors that achieves the same theoretical guarantees of its theoretical counterpart by replacing the unknown model parameters with the corresponding MAP estimate. Empirical studies confirm that our theoretical results closely match the remarkable performance observed in practice, even in finite, possibly small, sample regimes.
- Open Access
Concentration of discrepancy-based approximate Bayesian computation via Rademacher complexity
There has been an increasing interest on summary-free versions of approximate Bayesian computation (ABC), which replace distances among summaries with discrepancies between the empirical distributions of the observed data and the synthetic samples generated under the proposed parameter values. The success of these solutions has motivated theoretical studies on the limiting properties of the induced posteriors. However, current results (i) are often tailored to a specific discrepancy, (ii) require, either explicitly or implicitly, regularity conditions on the data generating process and the assumed statistical model, and (iii) yield bounds depending on sequences of control functions that are not made explicit. As such, there is the lack of a theoretical framework that (i) is unified, (ii) facilitates the derivation of limiting properties that hold uniformly, and (iii) relies on verifiable assumptions that provide concentration bounds clarifying which factors govern the limiting behavior of the ABC posterior. We address this gap via a novel theoretical framework that introduces the concept of Rademacher complexity in the analysis of the limiting properties for discrepancy-based ABC posteriors. This yields a unified theory that relies on constructive arguments and provides more informative asymptotic results and uniform concentration bounds, even in settings not covered by current studies. These advancements are obtained by relating the properties of summary-free ABC posteriors to the behavior of the Rademacher complexity associated with the chosen discrepancy within the family of integral probability semimetrics. This family extends summary-based ABC, and includes the Wasserstein distance and maximum mean discrepancy (MMD), among others. As clarified through a focus on the MMD case and via illustrative simulations, this perspective yields an improved understanding of summary-free ABC.
Open AccessConcentration of discrepancy-based approximate Bayesian computation via Rademacher complexity
There has been an increasing interest on summary-free versions of approximate Bayesian computation (ABC), which replace distances among summaries with discrepancies between the empirical distributions of the observed data and the synthetic samples generated under the proposed parameter values. The success of these solutions has motivated theoretical studies on the limiting properties of the induced posteriors. However, current results (i) are often tailored to a specific discrepancy, (ii) require, either explicitly or implicitly, regularity conditions on the data generating process and the assumed statistical model, and (iii) yield bounds depending on sequences of control functions that are not made explicit. As such, there is the lack of a theoretical framework that (i) is unified, (ii) facilitates the derivation of limiting properties that hold uniformly, and (iii) relies on verifiable assumptions that provide concentration bounds clarifying which factors govern the limiting behavior of the ABC posterior. We address this gap via a novel theoretical framework that introduces the concept of Rademacher complexity in the analysis of the limiting properties for discrepancy-based ABC posteriors. This yields a unified theory that relies on constructive arguments and provides more informative asymptotic results and uniform concentration bounds, even in settings not covered by current studies. These advancements are obtained by relating the properties of summary-free ABC posteriors to the behavior of the Rademacher complexity associated with the chosen discrepancy within the family of integral probability semimetrics. This family extends summary-based ABC, and includes the Wasserstein distance and maximum mean discrepancy (MMD), among others. As clarified through a focus on the MMD case and via illustrative simulations, this perspective yields an improved understanding of summary-free ABC.
- Open Access
Learning and forecasting of age-specific period mortality via B-spline processes with locally-adaptive dynamic coefficients
Although the analysis of human mortality has a well-established history, the attempt to accurately forecast future death-rate patterns for different age groups and time horizons still attracts active research. Such a predictive focus has motivated an increasing shift towards more flexible representations of age-specific period mortality trajectories at the cost of reduced interpretability. Although this perspective has led to successful predictive strategies, the inclusion of interpretable structures in modeling of human mortality can be, in fact, beneficial for improving forecasts. We pursue this direction via a novel B-spline process with locally-adaptive dynamic coefficients. Such a process outperforms state-of-the-art forecasting strategies by explicitly incorporating the core structures of period mortality within an interpretable formulation which enables inference on age-specific mortality trends and the corresponding rates of change across time. This is obtained by modeling the age-specific death counts via a Poisson log-normal model parameterized through a linear combination of B-spline bases with dynamic coefficients that characterize time changes in mortality rates via suitable stochastic differential equations. While flexible, the resulting formulation can be accurately approximated by a Gaussian state-space model that facilitates closed-form Kalman filtering, smoothing and forecasting, for both the trends of the spline coefficients and the corresponding first derivatives, which measure rates of change in mortality for different ages. As illustrated in applications to mortality data from different countries, our model outperforms state-of-the-art methods both in point forecasts and in calibration of predictive intervals. Moreover, it unveils substantial differences in mortality patterns across countries and ages, both in the past decades and during the COVID-19 pandemic.
Open AccessLearning and forecasting of age-specific period mortality via B-spline processes with locally-adaptive dynamic coefficients
Although the analysis of human mortality has a well-established history, the attempt to accurately forecast future death-rate patterns for different age groups and time horizons still attracts active research. Such a predictive focus has motivated an increasing shift towards more flexible representations of age-specific period mortality trajectories at the cost of reduced interpretability. Although this perspective has led to successful predictive strategies, the inclusion of interpretable structures in modeling of human mortality can be, in fact, beneficial for improving forecasts. We pursue this direction via a novel B-spline process with locally-adaptive dynamic coefficients. Such a process outperforms state-of-the-art forecasting strategies by explicitly incorporating the core structures of period mortality within an interpretable formulation which enables inference on age-specific mortality trends and the corresponding rates of change across time. This is obtained by modeling the age-specific death counts via a Poisson log-normal model parameterized through a linear combination of B-spline bases with dynamic coefficients that characterize time changes in mortality rates via suitable stochastic differential equations. While flexible, the resulting formulation can be accurately approximated by a Gaussian state-space model that facilitates closed-form Kalman filtering, smoothing and forecasting, for both the trends of the spline coefficients and the corresponding first derivatives, which measure rates of change in mortality for different ages. As illustrated in applications to mortality data from different countries, our model outperforms state-of-the-art methods both in point forecasts and in calibration of predictive intervals. Moreover, it unveils substantial differences in mortality patterns across countries and ages, both in the past decades and during the COVID-19 pandemic.
- 3 March 2023
- Published by Taylor & Francis in Journal of the American Statistical Association
- Vol. 118 (542) , 1451-1469
Bayesian Conjugacy in Probit, Tobit, Multinomial Probit and Extensions: A Review and New Results
A broad class of models that routinely appear in several fields can be expressed as partially or fully discretized Gaussian linear regressions. Besides including classical Gaussian response settings, this class also encompasses probit, multinomial probit and tobit regression, among others, thereby yielding one of the most widely-implemented families of models in routine applications. The relevance of such representations has stimulated decades of research in the Bayesian field, mostly motivated by the fact that, unlike for Gaussian linear regression, the posterior distribution induced by such models does not seem to belong to a known class, under the commonly-assumed Gaussian priors for the coefficients. This has motivated several solutions for posterior inference relying either on sampling-based strategies or on deterministic approximations that, however, still experience computational and accuracy issues, especially in high dimensions. The scope of this article is to review, unify and extend recent advances in Bayesian inference and computation for this core class of models. To address such a goal, we prove that the likelihoods induced by these formulations share a common analytical structure implying conjugacy with a broad class of distributions, namely the unified skew-normal (SUN), that generalize Gaussians to include skewness. This result unifies and extends recent conjugacy properties for specific models within the class analyzed, and opens new avenues for improved posterior inference, under a broader class of formulations and priors, via novel closed-form expressions, i.i.d. samplers from the exact SUN posteriors, and more accurate and scalable approximations from variational Bayes and expectation-propagation. Such advantages are illustrated in simulations and are expected to facilitate the routine-use of these core Bayesian models, while providing novel frameworks for studying theoretical properties and developing future extensions.
- 3 March 2023
- Published by Taylor & Francis in Journal of the American Statistical Association
- Vol. 118 (542) , 1451-1469
Bayesian Conjugacy in Probit, Tobit, Multinomial Probit and Extensions: A Review and New Results
A broad class of models that routinely appear in several fields can be expressed as partially or fully discretized Gaussian linear regressions. Besides including classical Gaussian response settings, this class also encompasses probit, multinomial probit and tobit regression, among others, thereby yielding one of the most widely-implemented families of models in routine applications. The relevance of such representations has stimulated decades of research in the Bayesian field, mostly motivated by the fact that, unlike for Gaussian linear regression, the posterior distribution induced by such models does not seem to belong to a known class, under the commonly-assumed Gaussian priors for the coefficients. This has motivated several solutions for posterior inference relying either on sampling-based strategies or on deterministic approximations that, however, still experience computational and accuracy issues, especially in high dimensions. The scope of this article is to review, unify and extend recent advances in Bayesian inference and computation for this core class of models. To address such a goal, we prove that the likelihoods induced by these formulations share a common analytical structure implying conjugacy with a broad class of distributions, namely the unified skew-normal (SUN), that generalize Gaussians to include skewness. This result unifies and extends recent conjugacy properties for specific models within the class analyzed, and opens new avenues for improved posterior inference, under a broader class of formulations and priors, via novel closed-form expressions, i.i.d. samplers from the exact SUN posteriors, and more accurate and scalable approximations from variational Bayes and expectation-propagation. Such advantages are illustrated in simulations and are expected to facilitate the routine-use of these core Bayesian models, while providing novel frameworks for studying theoretical properties and developing future extensions.