Count-weighted temporal networks often exhibit unequal dispersion in the edge weights, which cannot be fully explained by modelling observational heterogeneity through latent factors in the conditional mean. Therefore, we propose new dynamic network model classes exploiting the Generalized Poisson distribution to capture both under- and overdispersion. We consider three different dynamic specifications: latent factor dynamics, autoregressive dynamics, and latent position dynamics, and study some theoretical properties of the random networks, showing the impact of the dispersion parameter on the random network's connectivity. After discussing the parameter identification strategy, we present a Bayesian inference procedure along with a posterior sampling algorithm. A numerical illustration demonstrates the effectiveness of the designed algorithm and provides estimates of the misspecification bias when unequal dispersion is neglected. Our new models are then applied to two relevant dynamic datasets considered in previous studies: a set of bike-sharing dynamic networks and a set of dynamic media networks. Our results highlight the importance of explicitly modeling overdispersion for both an accurate in-sample fit and out-of-sample performance.

The integration of Artificial Intelligence (AI) within financial institutions has accelerated in response to mounting complexities and recurrent global crises, revealing the well-known but fair limitations of traditional economic models. Enhanced computational capacity and Big Data (BD) availability have enabled the adoption of Machine Learning (ML) across fields from banking, asset and risk management, and insurance, for tasks such as credit scoring, fraud detection, and market surveillance. AI facilitates portfolio construction and risk budgeting by synthesising historical and alternative data, while Natural Language Processing (NLP) aids in interpreting regulatory texts and central bank communications. However, the opacity of AI models introduces novel risks, including among others validation challenges and systemic vulnerabilities due to model convergence. When controlled appropriately, AI can paradoxically serve both as a risk source and a mitigation tool. This special issue presents rigorously peer-reviewed contributions that explore AI, ML, and BD applications in asset pricing, risk management, macroeconomic forecasting, and sustainable finance, offering valuable insights for academics, practitioners, and policymakers navigating financial uncertainty.

This paper investigates how the COVID-19 pandemic affected Italian firms’ expectations regarding future growth and prices, and the link between firm-level uncertainty and forecast errors. Using firm-level panel data from the Bank of Italy’s INVIND survey, we document that growth-expectation formation is stable across nearly three decades, including the COVID-19 pandemic, and that growth forecasts retain predictive content for realized outcomes in every year of the sample. Firms’ self-reported uncertainty about growth is informative about forecast accuracy in crisis years but not during 2020–2021, when uncertainty was driven by a common aggregate shock. At the aggregate level, cross-sectional disagreement among firms predicts the dispersion of subsequent realized outcomes, with predictability running only from expectations to outcomes. Workforce shortages are the dominant channel through which the pandemic disrupted firm expectations, more relevant than demand, supply, or financial channels. The price domain differs from real growth, with the response of expected prices to realized prices weakening during the late 1990s, the 2008–2009 financial crisis, and the 2022 inflation surge, while subjective price uncertainty remains a stable predictor of forecast errors.

This paper compares two approaches to constructing synthetic performance indicators for corporate default prediction: the expert-informed Synthetic Performance Indicator (ISP) and a Neural Network Synthetic Performance Indicator (NNISP). Both are built upon a common set of economic and financial indices from a large panel of firms operating in the Triveneto macro-region of Italy, including not only large companies but also small and medium enterprises (SMEs). The ISP relies on expert elicitation to assign weights to eight selected indicators, while the NNISP is obtained by calibrating the set of composition weights by training a Neural Network to predict a firm’s liquidity. We assess the predictive power of each score using multiple classification models and benchmark them against the full set of indicators. We show that the ISP offers stable and interpretable performance across model types, while the NNISP exhibits more flexibility in non-linear settings. These findings highlight a trade-off between expert-based interpretability and machine-driven adaptability, pointing to the potential of combined methodologies for future applications in credit risk modeling.

Analyzing and forecasting visitor flows has significant potential to enhance policy decisions, facilitate long- and short-term planning of local resources, and optimize tourism offerings. Leveraging Big Data in this context provides distinct advantages, particularly regarding real-time predictions. This paper proposes an econometric approach to account for uncertainty in the prediction model, a key source of forecasting errors. A description and prediction of daily visits in Shanghai City are provided using models commonly employed in tourist flows analysis. We show that model combination and selection procedures can serve as valuable tools for policymakers, equipping them with accurate forecasts to support tourist flow management.

A new dynamic latent space eigenmodel (LSM) is proposed for weighted temporal networks. The model accommodates integer-valued weights, excess of zeros, time-varying node positions (features), and time-varying network sparsity. The latent positions evolve according to a vector autoregressive process that accounts for lagged and contemporaneous dependence across nodes and features, a characteristic neglected in the LSM literature. A Bayesian approach is used to address two of the primary sources of inference intractability in dynamic LSMs: latent feature estimation and the choice of latent space dimension. We employ an efficient auxiliary-mixture sampler that performs data augmentation and supports conditionally conjugate prior distributions. A point-process representation of the network weights and the finite-dimensional distribution of the latent processes are used to derive a multi-move sampler in which each feature trajectory is drawn in a single block, without recursions. This sampling strategy is new to the network literature and can significantly reduce computational time while improving chain mixing. To avoid trans-dimensional samplers, a Laplace approximation of the partial marginal likelihood is used to design a partially collapsed Gibbs sampler. Overall, our procedure is general, as it can be easily adapted to static and dynamic settings, as well as to other discrete or continuous weight distributions.

We analyse fiscal policy in resource-rich economies using a novel Bayesian regime-switching panel model. The identified regimes capture pro- or countercyclical fiscal behaviour by allowing regime-specific shifts in the average fiscal stance, while the switches between the regimes have the interpretation of changes in fiscal policy. Applying the model to a panel of sixteen oil-producing economies, we show that fiscal policy has alternated between a procyclical and countercyclical regime multiple times over the sample. Furthermore, we find that fiscal policy is more volatile in the procyclical regime and that the probability of being in the procyclical regime is higher for OPEC countries than for non-OPEC countries. We also show that following either an increase or decrease in oil revenues, the growth in government expenditures is mostly increasing, suggesting an upward bias in expenditures in oil-producing countries. These are new findings in the literature.

Anomalies in economic and financial data – often linked to rare yet impactful events – are of theoretical interest, but can also severely distort inference. Although outlier-robust methodologies can be used, many researchers prefer pre-processing strategies that remove outliers. In this work, an efficient sequential Bayesian framework is proposed for outlier detection based on the predictive Bayes Factor (BF). The proposed method is specifically designed for large, multidimensional datasets and extends univariate Bayesian model outlier detection procedures to the matrix-variate setting. Leveraging power-discounted priors, tractable predictive BF are obtained, thereby avoiding computationally intensive techniques. The BF finite sample distribution, the test critical region, and robust extensions of the test are introduced by exploiting the sampling variability. The framework supports online detection with analytical tractability, ensuring both accuracy and scalability. Its effectiveness is demonstrated through simulations, and three applications to reference datasets in macroeconomics and finance are provided.

The availability of relational data can offer new insights into the functioning of the economy. Nevertheless, modeling the dynamics in network data with multiple types of relationships is still a challenging issue. Stochastic block models provide a parsimonious and flexible approach to network analysis. We propose a new stochastic block model for multidimensional networks, where layer-specific hidden Markov-chain processes drive the changes in community formation. The changes in the block membership of a node in a given layer may be influenced by its own past membership in other layers. This allows for clustering overlap, clustering decoupling, or more complex relationships between layers, including settings of unidirectional, or bidirectional, non-linear Granger block causality. We address the overparameterization issue of a saturated specification by assuming a Multi-Laplacian prior distribution within a Bayesian framework. Data augmentation and Gibbs sampling are used to make the inference problem more tractable. Through simulations, we show that standard linear models and the pairwise approach are unable to detect block causality in most scenarios. In contrast, our model can recover the true Granger causality structure. As an application to international trade, we show that our model offers a unified framework, encompassing community detection and Gravity equation modeling. We found new evidence of block Granger causality of trade agreements and flows and core-periphery structure in both layers on a large sample of countries.

A new integer-valued autoregressive process (INAR) with Generalised Lagrangian Katz (GLK) innovations is defined. This process family provides a flexible modeling framework for count data, allowing for under and over-dispersion, asymmetry, and excess of kurtosis. It also includes standard INAR models such as Generalized Poisson and Negative Binomial as special cases. It is shown that the GLK–INAR process is discrete semi-self-decomposable,​ infinite divisible, stable by aggregation and stationarity conditions are provided. Some extensions are discussed, such as the Markov-Switching and the zero-inflated​ GLK–INARs. A Bayesian inference framework and an efficient posterior approximation procedure are introduced. The proposed models are applied to 130 time series from Google Trend, which proxy the worldwide public concern about climate change. New evidence is found of heterogeneity across time, countries and keywords in the persistence, uncertainty, and long-run public awareness level.

News outlets are now more than ever incentivized to provide their audience with slanted news, while the intrinsic homophilic nature of online social media may exacerbate polarized opinions. Here, we propose a new dynamic latent space model for time-varying online audience-duplication networks, which exploits social media content to conduct inference on media bias and polarization of news outlets. Our model contributes to the literature in several directions: 1) we provide a model-embedded data-driven interpretation for the latent leaning of news outlets in terms of media bias; 2) we endow our model with Markov-switching dynamics to capture polarization regimes while maintaining a parsimonious specification; 3) we contribute to the literature on the statistical properties of latent space network models. The proposed model is applied to a set of data on the online activity of national and local news outlets from four European countries in the years 2015 and 2016. We find evidence of a strong positive correlation between our media slant measure and a well-grounded external source of media bias. In addition, we provide insight into the polarization regimes across the four countries considered.

A new flexible tensor model for multiple-equation regressions that accounts for latent regime changes is proposed. The model allows for dynamic coefficients and multi-dimensional covariates that vary across equations. The coefficients are driven by a common hidden Markov process that addresses structural breaks to enhance the model flexibility and preserve parsimony. A new soft PARAFAC hierarchical prior is introduced to achieve dimensionality reduction while preserving the structural information of the covariate tensor. The proposed prior includes a new multi-way shrinking effect to address over-parametrization issues while preserving interpretability and model tractability. Theoretical results are derived to help with the choice of the hyperparameters. An efficient Markov chain Monte Carlo (MCMC) algorithm based on random scan Gibbs and back-fitting strategy is designed with priority placed on computational scalability of the posterior sampling. The validity of the MCMC algorithm is demonstrated theoretically, and its computational efficiency is studied using numerical experiments in different parameter settings. The effectiveness of the model framework is illustrated using two original real data analyses. The proposed model exhibits superior performance compared to the current benchmark, Lasso regression.

This paper investigates the relationship between stock returns in the energy sector, energy uncertainty, and geopolitical risk. To this end, we propose a parsimonious and flexible model to extract common volatility factors (COVOL) from panel data. This general nonlinear multi-factor framework organizes panel units into groups based on different exposures to individual and compounding risks to reduce the number of parameters to be estimated. The group membership of the units is unknown, which naturally calls for using stochastic partition models. Random partition and compounding relationships are encoded in the weighted hyper-edges of a random hypergraph where the vertexes are the individual risks. In the empirical analysis, we study the volatility transmission in a multi-country setting and the role of individual and compounding risks.

A new flexible tensor-on-tenor regression model that accounts for latent regime changes is proposed. The coefficients are driven by a common hidden Markov process that addresses structural breaks to enhance the model flexibility and preserve parsimony. A new soft PARAFAC hierarchical prior is introduced to achieve dimensionality reduction while preserving the structural information of the covariate tensor. The proposed prior includes a new multi-way shrinking effect to address over-parametrization issues while preserving interpretability and model tractability. An efficient MCMC algorithm is introduced based on random scan Gibbs and back-fitting strategy. The model framework's effectiveness is illustrated using financial and commodity market volatility data. The proposed model exhibits superior performance compared to the current benchmark, Lasso regression.

The necessity of determining the probability of default often conflicts with the requirement for employing parsimonious methodologies. The inclusion of a large number of regressors in Logit models or Machine Learning approaches can lead to overfitting, thereby introducing biases that distort the results. This study aims to examine the extent to which an indicator that synthesizes information related to balance sheet metrics can achieve a performance comparable to that obtained through a comprehensive set of indicators. In this paper, we introduce the Synthetic Performance Indicator (ISP), which is derived from specific balance sheet indicators. We demonstrate its effectiveness as a synthetic measure of financial stability in localized settings. Furthermore, we assess its potential to serve as a viable alternative to the broader panel of indicators from which it is constructed. Finally, we provide further evidence of how Machine Learning approaches, despite being effective in-sample, perform poorly out-of-sample.

To address the common problem of high dimensionality in tensor regressions, we introduce a generalized tensor random projection method that embeds high-dimensional tensor-valued covariates into low-dimensional subspaces with minimal loss of information about the responses. The method is flexible, allowing for tensor-wise, mode-wise, or combined random projections as special cases. A Bayesian inference framework is provided featuring the use of a hierarchical prior distribution and a low-rank representation of the parameter. Strong theoretical support is provided for the concentration properties of the random projection and posterior consistency of the Bayesian inference. An efficient Gibbs sampler is developed to perform inference on the compressed data. To mitigate the sensitivity introduced by random projections, Bayesian model averaging is employed, with normalising constants estimated using reverse logistic regression. An extensive simulation study is conducted to examine the effects of different tuning parameters. Simulations indicate, and the real data application confirms, that compressed Bayesian tensor regression can achieve better out-of-sample prediction while significantly reducing computational cost compared to standard Bayesian tensor regression.

The techniques suggested in Frühwirth-Schnatter et al. (2024) concern sparsity and factor selection and have enormous potential beyond standard factor analysis applications. We show how these techniques can be applied to Latent Space (LS) models for network data. These models suffer from well-known identification issues of the latent factors due to likelihood invariance to factor translation, reflection, and rotation (see Hoff et al., 2002). A set of observables can be instrumental in identifying the latent factors via auxiliary equations (see Liu et al., 2021). These, in turn, share many analogies with the equations used in factor modeling, and we argue that the factor loading restrictions may be beneficial for achieving identification.

This paper introduces a new stochastic process with values in the set Z of integers with sign. The increments of process are Poisson differences and the dynamics has an autoregressive structure. We study the properties of the process and exploit the thinning representation to derive stationarity conditions and the stationary distribution of the process. We provide a Bayesian inference method and an efficient posterior approximation procedure based on Monte Carlo. Numerical illustrations on both simulated and real data show the effectiveness of the proposed inference.

The techniques suggested in Frühwirth-Schnatter et al. (2024) concern sparsity and factor selection and have enormous potential beyond standard factor analysis applications. We show how these techniques can be applied to Latent Space (LS) models for network data. These models suffer from well-known identification issues of the latent factors due to likelihood invariance to factor translation, reflection, and rotation (see Hoff et al., 2002). A set of observables can be instrumental in identifying the latent factors via auxiliary equations (see Liu et al., 2021). These, in turn, share many analogies with the equations used in factor modeling, and we argue that the factor loading restrictions may be beneficial for achieving identification.