# Research articles for the 2021-08-01

A data-science-driven short-term analysis of Amazon, Apple, Google, and Microsoft stocks
Shubham Ekapure,Nuruddin Jiruwala,Sohan Patnaik,Indranil SenGupta
arXiv

In this paper, we implement a combination of technical analysis and machine/deep learning-based analysis to build a trend classification model. The goal of the paper is to apprehend short-term market movement, and incorporate it to improve the underlying stochastic model. Also, the analysis presented in this paper can be implemented in a \emph{model-independent} fashion. We execute a data-science-driven technique that makes short-term forecasts dependent on the price trends of current stock market data. Based on the analysis, three different labels are generated for a data set: $+1$ (buy signal), $0$ (hold signal), or $-1$ (sell signal). We propose a detailed analysis of four major stocks- Amazon, Apple, Google, and Microsoft. We implement various technical indicators to label the data set according to the trend and train various models for trend estimation. Statistical analysis of the outputs and classification results are obtained.

Backtesting Systemic Risk Forecasts using Multi-Objective Elicitability
Tobias Fissler,Yannick Hoga
arXiv

Backtesting risk measure forecasts requires identifiability (for model validation) and elicitability (for model comparison). The systemic risk measures CoVaR (conditional value-at-risk), CoES (conditional expected shortfall) and MES (marginal expected shortfall), measuring the risk of a position $Y$ given that a reference position $X$ is in distress, fail to be identifiable and elicitable. We establish the joint identifiability of CoVaR, MES and (CoVaR, CoES) together with the value-at-risk (VaR) of the reference position $X$, but show that an analogue result for elicitability fails. The novel notion of multi-objective elicitability however, relying on multivariate scores equipped with an order, leads to a positive result when using the lexicographic order on $\mathbb{R}^2$. We establish comparative backtests of Diebold--Mariano type for superior systemic risk forecasts and comparable VaR forecasts, accompanied by a traffic-light approach. We demonstrate the viability of these backtesting approaches in simulations and in an empirical application to DAX 30 and S&P 500 returns.

Comparing cars with apples? Identifying the appropriate benchmark countries for relative ecological pollution rankings and international learning
Dominik Hartmann,Diogo Ferraz,Mayra Bezerra,Andreas Pyka,Flavio L. Pinheiro
arXiv

Research in Data Envelopment Analysis has created rankings of the ecological efficiency of countries' economies. At the same time, research in economic complexity has provided new methods to depict productive structures and has analyzed how economic diversification and sophistication affect environmental pollution indicators. However, no research so far has compared the ecological efficiency of countries with similar productive structures and levels of economic complexity, combining the strengths of both approaches. In this article, we use data on 774 different types of exports, CO2 emissions, and the ecological footprint of 99 countries to create a relative ecological pollution ranking (REPR). Moreover, we use methods from network science to reveal a benchmark network of the best learning partners based on country pairs with a large extent of export similarity, yet significant differences in pollution values. This is important because it helps to reveal adequate benchmark countries for efficiency improvements and cleaner production, considering that countries may specialize in substantially different types of economic activities. Finally, the article (i) illustrates large efficiency improvements within current global output levels, (ii) helps to identify countries that can best learn from each other, and (iii) improves the information base in international negotiations for the sake of a clean global production system.

Does foreign direct investment promote institutional development in Africa?
Fon, Roger,Filippaios, Fragkiskos,Stoian, Carmen,Lee, Soo-Hee
RePEC
Foreign direct investment (FDI) inflows into Africa have increased since the turn of the millennium, mainly due to FDI growth into African countries by multinational enterprises (MNEs) from developing economies. While African governments view this growth as a positive development for the continent, many governments in the West have raised concerns regarding the institutional impact of investments from developing economies. This paper examines the impact of FDI flows on institutional quality in African countries by distinguishing investments from developed versus developing economies. Previous empirical studies have found a significant relationship between FDI flows and institutional quality in African countries but regard the relationship as MNEs rewarding African countries for adopting institutional reforms. However, little attention has been paid to the reverse causality, i.e. that FDI can cause an institutional change in African countries. Using bilateral greenfield FDI flows between 56 countries during 2003-2015, we find no significant FDI effect from developed and developing economies on institutional quality in host countries. However, aggregate FDI flows from developed and developing economies have a significant positive effect on host country institutional quality but differ concerning the impact's timing. In contrast, we find no significant effect of FDI flows from China on host country institutional quality. Our results are robust to alternative measures of institutional quality.

Financial intermediation and risk in decentralized lending protocols
Carlos Castro-Iragorri,Julian Ramirez,Sebastian Velez
arXiv

We provide an overview of decentralized protocols like Compound and Aave that offer collateralized loans for cryptoasset investors. Compound and Aave are two of the most important application in the decentralized finance (DeFi) ecosystem. Using publicly available information on rates, supply and borrow activity, and accounts we analyze different elements of the protocols. In particular, we estimate ex-post margins that give a comprehensive account of the cost of financial intermediation. We find that ex-post margins considering all markets are 1% and lower for stablecoin markets. In addition, we estimate quarterly indicators regarding solvency, asset quality, earnings and market risk similar to the ones used in traditional banking. This provides a first look at the use of these metrics and a comparison between the similarities and challenges to our understanding of financial intermediation in these protocols based on tools used for traditional banking.

International Trade Network: Country centrality and COVID-19 pandemic
Roberto Antonietti,Paolo Falbo,Fulvio Fontini,Rosanna Grassi,Giorgio Rizzini
arXiv

International trade is based on a set of complex relationships between different countries that can be modelled as an extremely dense network of interconnected agents. On the one hand, this network might favour the economic growth of countries, but on the other, it can also favour the diffusion of diseases, like the COVID-19. In this paper, we study whether, and to what extent, the topology of the trade network can explain the rate of COVID-19 diffusion and mortality across countries. We compute the countries' centrality measures and we apply the community detection methodology based on communicability distance. Then, we use these measures as focal regressors in a negative binomial regression framework. In doing so, we also compare the effect of different measures of centrality. Our results show that the number of infections and fatalities are larger in countries with a higher centrality in the global trade network.

Propaganda, Alternative Media, and Accountability in Fragile Democracies
Anqi Li,Davin Raiha,Kenneth W. Shotts
arXiv

We develop a model of electoral accountability with mainstream and alternative media. In addition to regular high- and low-competence types, the incumbent may be an aspiring autocrat who controls the mainstream media and will subvert democracy if retained in office. A truthful alternative media can help voters identify and remove these subversive types while re-electing competent leaders. A malicious alternative media, in contrast, spreads false accusations about the incumbent and demotivates policy effort. If the alternative media is very likely be malicious and hence is unreliable, voters ignore it and use only the mainstream media to hold regular incumbents accountable, leaving aspiring autocrats to win re-election via propaganda that portrays them as effective policymakers. When the alternative media's reliability is intermediate, voters heed its warnings about subversive incumbents, but the prospect of being falsely accused demotivates effort by regular incumbents and electoral accountability breaks down.

Reconciling revealed and stated measures for willingness to pay in recreation by building a probability model
Edoh Y. Amiran,Joni S. James Charles
arXiv

The consumers' willingness to pay plays an important role in economic theory and in setting policy. For a market, this function can often be estimated from observed behavior -- preferences are revealed. However, economists would like to measure consumers' willingness to pay for some goods where this can only be measured through stated valuation. Confirmed convergence of valuations based on stated preferences as compared to valuations based on revealed preferences is rare, and it is important to establish circumstances under which one can expect such convergence. By building a simple probabilistic model for the consumers' likelihood of travel, we provide an approach that should make comparing stated and revealed preferences easier in cases where the preference is tied to travel or some other behavior whose cost can be measured. We implemented this approach in a pilot study and found an estimate of willingness to pay for visiting an environmentally enhanced recreational site based on actual travel in good agreement with an estimate based on a survey using stated preferences. To use the probabilistic model we used population statistics to adjust for the relevant duration and thus compare stated and revealed responses.

Relational Graph Neural Networks for Fraud Detection in a Super-App environment
Jaime D. Acevedo-Viloria,Luisa Roa,Soji Adeshina,Cesar Charalla Olazo,Andrés Rodríguez-Rey,Jose Alberto Ramos,Alejandro Correa-Bahnsen
arXiv

Large digital platforms create environments where different types of user interactions are captured, these relationships offer a novel source of information for fraud detection problems. In this paper we propose a framework of relational graph convolutional networks methods for fraudulent behaviour prevention in the financial services of a Super-App. To this end, we apply the framework on different heterogeneous graphs of users, devices, and credit cards; and finally use an interpretability algorithm for graph neural networks to determine the most important relations to the classification task of the users. Our results show that there is an added value when considering models that take advantage of the alternative data of the Super-App and the interactions found in their high connectivity, further proofing how they can leverage that into better decisions and fraud detection strategies.

Spousal Occupational Sorting and COVID-19 Incidence: Evidence from the United States
Egor Malkov
arXiv

How do matching of spouses and the nature of work jointly shape the distribution of COVID-19 health risk? To address this question, I study the association between the incidence of COVID-19 and the degree of spousal sorting into occupations that differ by contact intensity at the workplace. The mechanism, that I explore, implies that the higher degree of positive spousal sorting mitigates intra-household contagion and this translates into smaller number of individuals exposed to COVID-19 risk. Using the U.S. data at the state level, I argue that spousal sorting is an important factor for understanding the disparities in the prevalence of COVID-19 during the early stages of the pandemic. First, I document that it creates about two thirds of the U.S. dual-earner couples that are exposed to higher COVID-19 health risk due to within-household transmission. Moreover, I uncover a substantial heterogeneity in the degree of spousal sorting by state. Next, for the first week of April 2020, I estimate that a one standard deviation increase in the measure of spousal sorting is associated with a 30% reduction in the total number of cases per 100000 inhabitants, and a 39.3% decline in the total number of deaths per 100000 inhabitants. Furthermore, I find substantial temporal heterogeneity as the coefficients decline in magnitude over time. My results speak to the importance of policies that allow to mitigate intra-household contagion.

The Adaptive Multi-Factor Model and the Financial Market
Liao Zhu
arXiv

Modern evolvements of the technologies have been leading to a profound influence on the financial market. The introduction of constituents like Exchange-Traded Funds, and the wide-use of advanced technologies such as algorithmic trading, results in a boom of the data which provides more opportunities to reveal deeper insights. However, traditional statistical methods always suffer from the high-dimensional, high-correlation, and time-varying instinct of the financial data. In this dissertation, we focus on developing techniques to stress these difficulties. With the proposed methodologies, we can have more interpretable models, clearer explanations, and better predictions.

The characteristic function of Gaussian stochastic volatility models: an analytic expression
Eduardo Abi Jaber
arXiv

Stochastic volatility models based on Gaussian processes, like fractional Brownian motion, are able to reproduce important stylized facts of financial markets such as rich autocorrelation structures, persistence and roughness of sample paths. This is made possible by virtue of the flexibility introduced in the choice of the covariance function of the Gaussian process. The price to pay is that, in general, such models are no longer Markovian nor semimartingales, which limits their practical use. We derive, in two different ways, an explicit analytic expression for the joint characteristic function of the log-price and its integrated variance in general Gaussian stochastic volatility models. Such analytic expression can be approximated by closed form matrix expressions. This opens the door to fast approximation of the joint density and pricing of derivatives on both the stock and its realized variance using Fourier inversion techniques. In the context of rough volatility modeling, our results apply to the (rough) fractional Stein--Stein model and provide the first analytic formulae for option pricing known to date, generalizing that of Stein--Stein, Sch{\"o}bel-Zhu and a special case of Heston.