【Abstract】Financial technologies, boosted by the availability of machine learning models, are expanding in all areas of finance: from payments (peer to peer lending) to asset management (robot advisors) to payments (blockchain coins). Machine learning models typically achieve a high accuracy at the expense of an insufficient explainability. Moreover, according to the proposed regulations, high-risk AI applications based on machine learning must be "trustworthy", and comply with a set of mandatory requirements, such as Sustainability and Fairness. To date there are no standardised metrics that can ensure an overall assessment of the trustworthiness of AI applications in finance. To fill the gap, we propose a set of integrated statistical methods, based on the Lorenz Zonoid tool, that can be used to assess and monitor over time whether an AI application is trustworthy. Specifically, the methods will measure Sustainability (in terms of robustness with respect to anomalous data), Accuracy (in terms of predictive accuracy), Fairness (in terms of prediction bias across different population groups) and Explainability (in terms of human understanding and oversight). We apply our proposal to an easily downloadable dataset, that concerns financial prices, to make our proposal easily reproducible.
【Abstract】This study examines whether the emotions contained in new photos can affect the cryptocurrency market. Utilizing the daily data on top 100 cryptocurrencies, we find that the surge in ratio of photos comprising pessimistic tones is associated with negative coin returns. Photo sentiment positively predicts subsequent returns and trading intensity, implying the subsequent corrections. The photo sentiment also drives risks up with higher price volatilities. The predictive power is more pronounced during periods of elevated fear proxied by investors' risk aversion. Our results remain robust with alternative sentiment proxies, risk-adjusted returns, and a battery of subsample analyses.
【Abstract】Usually, hypotheses involving cooperative interactions and continuous interplay among interacting agents may result in somewhat conservatism, drastically hurdling potential about interpretations of prevalent aggregation phenomena arising in both nature and practical application. Of interest uncovered in this article is the cooperative-competitive consensus problems for multi-agent systems by leverage of impulsive-based control perspective. We first discuss the possibility of quantifying interaction relation of the agents and the description of cooperative-competitive phenomena that capture massive interest recently. Specifically, the former is done by algebraic graph theory, following the same research avenue as the convention; while the cooperative-competitive information is described by some nonzero scaling parameter. We shall pursue the tie that enables to anchor the consensus of participating agents, without depending on the signed graph theory. Subsequently, a manner relying on the local information is carried out to induce the consensus error to circumvent the global information. Then, consensus condition, in connection with the features of both continuous and impulsive ߚbased dynamics, is obtained, as well as several typical circumstances are also elaborated. Both the proposed consensus algorithms and the derived results are underpinned via a numerical study eventually.
【Abstract】Social media has become an outlet for extremists to fundraise and organize. While governments deliberate on how to regulate, some social media companies have removed creators of offensive content -deplatforming. I estimate the effects of deplatforming on revenue and viewership, using variation in the timing of removals across two video-streaming companies - YouTube, and its far-right competitor, Bitchute. Being deplatformed on Youtube results in a 30% increase in weekly Bitcoin revenue and a 50% increase in viewership on Bitchute. This increase in Bitchute activity is less than that on YouTube, meaning that deplatforming works in decreasing a content creator's overall views and revenue.
【Abstract】This paper analyzes the response of cryptocurrency returns to the movement of economic policy uncertainty (EPU) and stock market volatility (VIX), as well as a few macroeconomic variables: gold price, interest rate, inflation rate, and oil price. Vector error correction model and regression model are applied to examine the linkage between these variables using data from 2015 to 2022. The analysis reveals that the selected variables have a positive and significant impact on cryptocurrency returns. This suggests that cryptocurrency can be considered a safe haven for investment. The paper also suggests a number of policies to ensure the protection of investment, control money supply and stock market instability, stabilize economic uncertainty, and systematize economic variables. This paper advocates a well-connected network and active participation of stakeholders such as government, central bank, security exchange, and financial institutions will help to streamline irrational movements and enhance the acceptability of cryptocurrencies through the framing and implementation of necessary regulations.
