John Cabot University ScholarShip

The information age of big data has increased the use of predictive data analytics for decision-making, leading to complex societal and legal debates with respect to statistical discrimination and the protection of civil liberties. This paper presents a simple pedagogic narrative in the context of asymmetric information. Students walk through a predictive-algorithmic decision-making process by a principal and agent. A simple predictive algorithm is constructed that leads to statistical discrimination and incentivizes principals to acquire more information confronting civil liberties of the right to privacy and protection from unnecessary surveillance. The lesson can be applied to labor, financial, and insurance markets.

This research investigates the macroeconomic determinants of crypto adoption, illuminating the potentials of cryptocurrencies to accelerate financial inclusion. By exploiting an extensive dataset from 165 countries between 2019 and 2021, this study employs various econometric methodologies, including Panel Feasible Generalized Least Squares (PFGLS), Robust Least Squares (RLS), and Quantile Regressions (QR). These classic econometric techniques are complemented by several machine learning techniques such as Bagging, Boosting, and Support Vector Machine (SVM) regressions, as well as Artificial Neural Networks (ANNs) and Naïve Bayes (NB) classification algorithms. The results show an interesting trend: cryptocurrency adoption is more prevalent in countries with robust financial markets and higher education levels. This suggests that crypto adoption is primarily a byproduct of sophisticated financial environments and an educated population, rather than a direct facilitator of financial inclusion.

The aim of this paper is to estimate the effects of the COVID-19 pandemic on the wage polarization in Italy, combining individual characteristics with their task content in terms of physical proximity within the workplace. We use an innovative dataset which combines data from two sample surveys, the Italian Labor Force Survey and Italian Survey of Professions, which provides information on nature and content of the tasks. First, by employing a non-parametric method (the Relative Distribution) we detect a general increasing wage polarization in the sub-period 2020–2019, driven by lowest deciles, after a reduction in the previous one (2019–10). Different groups have been also isolated. Workers with low education, high proximity to customers job, such as the immigrant, younger and female ones are the categories that more suffered the general downgrading of the Italian wages happened during the COVID-19 crisis.

In the race to harvest the power of Artificial Intelligence (AI) in virtually every field, researchers and practitioners are faced with an ever increasing supply of novel tools that have not undergone domain-specific tests. This paper informs the methodological choices of researchers in economics and finance by comparing the performance of three Natural Language Processing (NLP) methods at an important task, namely using text analysis for portfolio diversification. Portfolio management can benefit from analysing text data in the form of company descriptions, since the returns of companies with similar descriptions tend to be correlated and consequently, portfolios of dissimilar companies should have lower risk. In this paper, three NLP methods are used to construct so-called minimum semantic concentration portfolios, which are designed to leverage the semantic diversity of the business descriptions of constituent companies to reduce portfolio volatility. Two widely used large language models (BERT and GPT) and an alternative AI solution inspired by neuroscience, called semantic fingerprinting are put to the test of comparing meaningfully the business descriptions of the S&P 500 and respectively Europe 600 constituents in order to derive actionable investment insights. The results show that all three NLP methods are able to extract relevant information from company descriptions: the minimum semantic concentration portfolios have significantly lower volatility than portfolios constructed with randomly chosen weights. While no NLP method is able to claim absolute superiority over its peers, semantic fingerprinting appears the most consistent and robust performer, since BERT and GPT demonstrate not only their potential but also a caveat, as their performances are volatile even across very similar tasks.