Journal of Finance and Data Science最新文献

Startups are a key force driving economic development, and the success of these high-risk ventures can bring huge profits to venture capital firms. The ability to predict the success of startups is a major advantage for investors to outperform their competitors. In this study, we explore the potential of using publicly available LinkedIn profiles as an alternative and complementary data source to Crunchbase for predicting startup success. We provide a comprehensive review of the existing literature on the factors that influence startup success to create a large set of features for predictive modeling. We train two models for predicting startup success employing light gradient boosting that use LinkedIn data as a standalone and as a complementary data source, and compare them to baseline models based on Crunchbase data. We show that using LinkedIn as a complementary data source yields the best result with a mean area under the curve (AUC) value of 84%. We also provide a thorough analysis of what types of information contribute most to modeling startup success using the Shapley value method. Our models and analysis can be used to develop a decision support system to facilitate startup screening and the due diligence process for venture capital firms.

We study the relation between bond characteristics and corporate bond returns in China's two distinct and segmented bond markets—the interbank market and the exchange market—with a large cross-sectional dataset of 8318 corporate bonds from January 2010 to December 2022. Corporate bonds with large sizes, long maturities, old ages, poor credit ratings and large Amihud illiquidity earn high monthly returns in the interbank market. The return predictive patterns of bond size, time to maturity, and credit rating are similar in the exchange market, but bond age and Amihud illiquidity predict returns in the opposite direction, implying market segmentation. We construct two factors based on credit rating and Amihud illiquidity to represent the common risk of corporate bonds—credit risk and liquidity risk—and use the Hansen-Jagannathan distance to evaluate the performances of factors in explaining the returns of corporate bond portfolios. We find that the two characteristic-based factors help reduce the model specification errors of the five factors in Fama and French (1993).

Chengtou bond is the only asset with market prices that can capture the funding cost of Chinese local government debt. In contrast to the U.S. municipal bonds, Chengtou bonds are issued by private corporations but implicitly guaranteed by local and the central governments, which are reflected by novel risk characteristics—real estate GDP and political risk. One standard deviation increase in local real estate GDP (political risk) corresponds to 10 (9) basis points decrease (increase) in bond yields, respectively. However, conditional on political risk, real estate GDP actually increases bond yields, suggesting that only local governments with low political risk can enjoy the low funding costs driven by high real estate growth.

This paper introduces the minCluster portfolio, which is a portfolio optimization method combining the optimization of downside risk measures, hierarchical clustering and cellwise robustness. Using cellwise robust association measures, the minCluster portfolio is able to retrieve the underlying hierarchical structure in the data. Furthermore, it provides downside protection by using tail risk measures for portfolio optimization. We show through simulation studies and a real data example that the minCluster portfolio produces better out-of-sample results than mean-variances or other hierarchical clustering based approaches. Cellwise outlier robustness makes the minCluster method particularly suitable for stable optimization of portfolios in highly volatile markets, such as portfolios containing cryptocurrencies.

China's markets gained 3.86% around December 4, 2012, when the Party announced anti-corruption reforms. State-owned enterprises (SOEs) with higher past entertainment and travel costs (ETC) gained more. NonSOEs gained in more liberalized provinces, especially those with high past ETC, productivity, growth opportunities, and external financing. NonSOEs lost in the least liberalized provinces, especially those with high past ETC. These findings support investors' expect reduced official corruption to create value overall, reduce SOE waste, lower bureaucratic barriers to efficient resource allocation where markets function, and impede business in unliberalized provinces, where “getting things done” still requires investment in greasing bureaucratic gears.

The paper aims to indentify and compare the determinants of the overall FinTech market expansion and its major segments – cryptocurrency and peer-to-peer lending markets – in a dataset, which covers 64 countries and 51 potentially relevant factors. To this end, we apply a battery of state-of-the-art variable selection techniques from machine learning, comprising Bayesian model averaging (BMA), least absolute shrinkage and selection operator (LASSO), variable selection using random forests (VSURF) as well as spike-and-slab regression. We document substantial heterogeneity of the pivotal determinants across the FinTech market as a whole and its major segments. Thus, specific rather than general policy measures are needed to foster the development of standalone FinTech market segments. Moreover, our findings suggest that most countries don't need to seek a universal specialization in FinTech activities, concentrating on the segment where they have a competitive edge in terms of the pivotal determinants which drive its expansion.

We use supervised learning to identify factors that predict the cross-section of returns and maximum drawdown for stocks in the US equity market. Our data run from January 1970 to December 2019 and our analysis includes ordinary least squares, penalized linear regressions, tree-based models, and neural networks. We find that the most important predictors tended to be consistent across models, and that non-linear models had better predictive power than linear models. Predictive power was higher in calm periods than in stressed periods. Environmental, social, and governance indicators marginally impacted the predictive power of non-linear models in our data, despite their negative correlation with maximum drawdown and positive correlation with returns. Upon exploring whether ESG variables are captured by some models, we find that ESG data contribute to the prediction nonetheless.

Climate change and sustainability have become societal focal points in the last decade. Consequently, companies have been increasingly characterized by non-financial information, such as environmental, social, and governance (ESG) scores, based on which companies can be grouped into ESG classes. While many scholars have questioned the relationship between financial performance and risks of assets belonging to different ESG classes, the question about dependence among ESG classes is still open. Here, we focus on understanding the dependence structures of different ESG class indices and the market index through the lens of copula models. After a thorough introduction to vine copula models, we explain how cross-sectional and temporal dependencies can be captured by models based on vine copulas, more specifically, using ARMA-GARCH and stationary vine copula models. Using real-world ESG data over a long period with different economic states, we find that assets with medium ESG scores tend to show weaker dependence to the market, while assets with extremely high or low ESG scores tend to show stronger, non-Gaussian dependence.

In this paper, we document the importance of memory in machine learning (ML)-based models relying on firm characteristics for asset pricing. We find that predictive algorithms perform best when they are trained on long samples, with long-term returns as dependent variables. In addition, we report that persistent features play a prominent role in these models. When applied to portfolio choice, we find that investors are always better off predicting annual returns, even when rebalancing at lower frequencies (monthly or quarterly). Our results remain robust to transaction costs and risk scaling, thus providing useful indications to quantitative asset managers.