A new data integration framework for Covid-19 social media information.

The Covid-19 pandemic presents a serious threat to people's health, resulting in over 250 million confirmed cases and over 5 million deaths globally. To reduce the burden on national health care systems and to mitigate the effects of the outbreak, accurate modelling and forecasting methods for short- and long-term health demand are needed to inform government interventions aiming at curbing the pandemic. Current research on Covid-19 is typically based on a single source of information, specifically on structured historical pandemic data. Other studies are exclusively focused on unstructured online retrieved insights, such as data available from social media. However, the combined use of structured and unstructured information is still uncharted. This paper aims at filling this gap, by leveraging historical and social media information with a novel data integration methodology. The proposed approach is based on vine copulas, which allow us to exploit the dependencies between different sources of information. We apply the methodology to combine structured datasets retrieved from official sources and a big unstructured dataset of information collected from social media. The results show that the combined use of official and online generated information contributes to yield a more accurate assessment of the evolution of the Covid-19 pandemic, compared to the sole use of official data.

Bayesian analysis of immigration in Europe with generalized logistic regression.

The number of immigrants moving to and settling in Europe has increased over the past decade, making migration one of the most topical and pressing issues in European politics. It is without a doubt that immigration has multiple impacts, in terms of economy, society and culture, on the European Union. It is fundamental to policy-makers to correctly evaluate people's attitudes towards immigration when designing integration policies. Of critical interest is to properly discriminate between subjects who are favourable towards immigration from those who are against it. Public opinions on migration are typically coded as binary responses in surveys. However, traditional methods, such as the standard logistic regression, may suffer from computational issues and are often not able to accurately model survey information. In this paper we propose an efficient Bayesian approach for modelling binary response data based on the generalized logistic regression. We show how the proposed approach provides an increased flexibility compared to traditional methods, due to its ability to capture heavy and light tails. The power of our methodology is tested through simulation studies and is illustrated using European Social Survey data on immigration collected in different European countries in 2016-2017.

Analysis of paediatric visual acuity using Bayesian copula models with sinh-arcsinh marginal densities.

We analyse paediatric ophthalmic data from a large sample of children aged between 3 and 8 years. We use a Bayesian additive conditional bivariate copula regression model with sinh-arcsinh marginal densities with location, scale, and shape parameters that depend smoothly on a covariate. We perform Bayesian inference about the unknown quantities of our model using a specially tailored Markov chain Monte Carlo algorithm. We gain new insights about the processes, which determine transformations in visual acuity with respect to age, including the nature of joint changes in both eyes as modelled with the age-related copula dependence parameter. We analyse posterior predictive distributions to identify children with unusual sight characteristics, distinguishing those who are bivariate, but not univariate outliers. In this way, we provide an innovative tool that enables clinicians to identify children with unusual sight who may otherwise be missed. We compare our simultaneous Bayesian method with a two-step frequentist generalised additive modelling approach.

Exploring the Potential for Multivariate Fragility Representations to Alter Flood Risk Estimates.

In flood risk analysis, limitations in the multivariate statistical models adopted to model the hydraulic load have restricted the probability of a defense suffering structural failure to be expressed conditionally on a single hydraulic loading variable. This is an issue at the coastal level where multiple loadings act on defenses with the exact combination of loadings dictating their failure probabilities. Recently, a methodology containing a multivariate statistical model with the flexibility to robustly capture the dependence structure between the individual loadings was used to derive extreme nearshore loading conditions. Its adoption will permit the incorporation of more precise representations of a structure's vulnerability in future analyses. In this article, a fragility representation of a shingle beach, where the failure probability is expressed over a three-dimensional loading parameter space-water level, wave height, and period-is derived at two localities. Within the approach, a Gaussian copula is used to capture any dependencies between the simplified geometric parameters of a beach's shape. Beach profiles are simulated from the copula and the failure probability, given the hydraulic load, determined by the reformulated Bradbury barrier inertia parameter model. At one site, substantial differences in the annual failure probability distribution are observed between the new and existing approaches. At the other, the beach only becomes vulnerable after a significant reduction of the crest height with its mean annual failure probability close to that presently predicted. It is concluded that further application of multivariate approaches is likely to yield more effective flood risk management.