Pub Date : 2025-01-29DOI: 10.1016/j.jfds.2025.100149
Marcel Lee , Alan Spark
Stock returns are moved by many risk factors. Thematic stock indices try to represent these factors, but are limited by the fact that risk factors are not directly observable. This paper introduces a method to uncover hidden risk factors through text analysis. It applies the dynamic variant of the Latent Dirichlet Allocation (LDA) model to annual and quarterly reports to find a topic distribution for each stock. This is then interpreted as the risk factor partition and transformed into a standard normal basis which corresponds to pure risk factors. The weights indicate the proportions necessary to combine the equities into tradable topic indices. The need for human intervention is minimized by determining the optimal parameters automatically.
{"title":"Unsupervised generation of tradable topic indices through textual analysis","authors":"Marcel Lee , Alan Spark","doi":"10.1016/j.jfds.2025.100149","DOIUrl":"10.1016/j.jfds.2025.100149","url":null,"abstract":"<div><div>Stock returns are moved by many risk factors. Thematic stock indices try to represent these factors, but are limited by the fact that risk factors are not directly observable. This paper introduces a method to uncover hidden risk factors through text analysis. It applies the dynamic variant of the <em>Latent Dirichlet Allocation</em> (LDA) model to annual and quarterly reports to find a topic distribution for each stock. This is then interpreted as the risk factor partition and transformed into a standard normal basis which corresponds to pure risk factors. The weights indicate the proportions necessary to combine the equities into tradable topic indices. The need for human intervention is minimized by determining the optimal parameters automatically.</div></div>","PeriodicalId":36340,"journal":{"name":"Journal of Finance and Data Science","volume":"11 ","pages":"Article 100149"},"PeriodicalIF":0.0,"publicationDate":"2025-01-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143454732","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-01-10DOI: 10.1016/j.jfds.2025.100151
Helge Holden , Lars Holden
The increasing fraction of passive funds influences stock market variability since passive investors behave differently than active investors. We demonstrate via simulations how portfolios that rebalance between different classes of assets influence the market variability. We prove that the optimal strategy for such portfolios when we include transaction costs, is only to rebalance when the portfolio leaves a no-trade region in the state space. This is the case also when the expectation and volatility of the prices are inhomogeneous. We show that portfolios that apply an optimal rebalance strategy reduce the variability in the stock market measured in the sum of the distances between local minimum and maximum of the prices in the stock market, also when these portfolios constitute only a small part of the market. However, the more usual rebalance strategies that only consider to rebalance at the end of a month or a quarter, have a much weaker influence on the market variability.
{"title":"Optimal rebalancing strategies reduce market variability","authors":"Helge Holden , Lars Holden","doi":"10.1016/j.jfds.2025.100151","DOIUrl":"10.1016/j.jfds.2025.100151","url":null,"abstract":"<div><div>The increasing fraction of passive funds influences stock market variability since passive investors behave differently than active investors. We demonstrate via simulations how portfolios that rebalance between different classes of assets influence the market variability. We prove that the optimal strategy for such portfolios when we include transaction costs, is only to rebalance when the portfolio leaves a no-trade region in the state space. This is the case also when the expectation and volatility of the prices are inhomogeneous. We show that portfolios that apply an optimal rebalance strategy reduce the variability in the stock market measured in the sum of the distances between local minimum and maximum of the prices in the stock market, also when these portfolios constitute only a small part of the market. However, the more usual rebalance strategies that only consider to rebalance at the end of a month or a quarter, have a much weaker influence on the market variability.</div></div>","PeriodicalId":36340,"journal":{"name":"Journal of Finance and Data Science","volume":"11 ","pages":"Article 100151"},"PeriodicalIF":0.0,"publicationDate":"2025-01-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143173405","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-01-09DOI: 10.1016/j.jfds.2025.100150
Xiangwu Zuo, Anxiao (Andrew) Jiang
Symbolic Regression is a machine learning technique that discovers an unknown function from its samples. Compared to conventional regression techniques (e.g., linear regression, polynomial regression, etc.), Symbolic Regression does not limit the discovered function to specific forms (e.g., linear functions, polynomials, etc.). Its recent developments are enabling its application to various fields, including both scientific study and engineering research. However, in spite of its flexibility, Symbolic Regression still faces one limitation: given datasets from different systems in the same domain, Symbolic Regression needs to find a distinct function for each dataset, instead of finding a more general yet succinct function that can fit all the datasets through the adjustments of its coefficients. The latter approach, which is termed “Symbolic Modeling” in this work, can be seen as a generalization of Symbolic Regression and has important applications to both academia and industry. This work elucidates Symbolic Modeling and unveils a cutting-edge algorithm, deriving its principles from deep learning and genetic programming. This algorithm is implemented into an application, showcasing its practical utility in the field of financial asset pricing, an integral facet of finance that concentrates on asset valuation. It is shown that Symbolic Modeling compares favorably to existing asset pricing models in multiple aspects.
{"title":"Symbolic Modeling for financial asset pricing","authors":"Xiangwu Zuo, Anxiao (Andrew) Jiang","doi":"10.1016/j.jfds.2025.100150","DOIUrl":"10.1016/j.jfds.2025.100150","url":null,"abstract":"<div><div>Symbolic Regression is a machine learning technique that discovers an unknown function from its samples. Compared to conventional regression techniques (e.g., linear regression, polynomial regression, <em>etc.</em>), Symbolic Regression does not limit the discovered function to specific forms (e.g., linear functions, polynomials, <em>etc.</em>). Its recent developments are enabling its application to various fields, including both scientific study and engineering research. However, in spite of its flexibility, Symbolic Regression still faces one limitation: given datasets from different systems in the same domain, Symbolic Regression needs to find a distinct function for each dataset, instead of finding a more general yet succinct function that can fit all the datasets through the adjustments of its coefficients. The latter approach, which is termed “Symbolic Modeling” in this work, can be seen as a generalization of Symbolic Regression and has important applications to both academia and industry. This work elucidates Symbolic Modeling and unveils a cutting-edge algorithm, deriving its principles from deep learning and genetic programming. This algorithm is implemented into an application, showcasing its practical utility in the field of financial asset pricing, an integral facet of finance that concentrates on asset valuation. It is shown that Symbolic Modeling compares favorably to existing asset pricing models in multiple aspects.</div></div>","PeriodicalId":36340,"journal":{"name":"Journal of Finance and Data Science","volume":"11 ","pages":"Article 100150"},"PeriodicalIF":0.0,"publicationDate":"2025-01-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143173404","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-12-01DOI: 10.1016/j.jfds.2024.100146
Minwu Kim, Sidahmend Benahderrahmane, Talal Rahwan
This research presents a predictive model to identify potential targets of activist investment funds—entities that acquire significant corporate stakes to influence strategic and operational decisions, ultimately enhancing shareholder value. Predicting such targets is crucial for companies aiming to mitigate intervention risks, activist funds seeking optimal investments, and investors looking to leverage potential stock price gains. Using data from the Russell 3000 index from 2016 to 2022, we evaluated 123 model configurations incorporating diverse imputation, oversampling, and machine learning techniques. Our best model achieved an AUC-ROC of 0.782, demonstrating its capability to effectively predict activist fund targets. To enhance interpretability, we employed the Shapley value method to identify key factors influencing a company’s likelihood of being targeted, highlighting the dynamic mechanisms underlying activist fund target selection. These insights offer a powerful tool for proactive corporate governance and informed investment strategies, advancing understanding of the mechanisms driving activist investment decisions.
{"title":"Interpretable machine learning model for predicting activist investment targets","authors":"Minwu Kim, Sidahmend Benahderrahmane, Talal Rahwan","doi":"10.1016/j.jfds.2024.100146","DOIUrl":"10.1016/j.jfds.2024.100146","url":null,"abstract":"<div><div>This research presents a predictive model to identify potential targets of activist investment funds—entities that acquire significant corporate stakes to influence strategic and operational decisions, ultimately enhancing shareholder value. Predicting such targets is crucial for companies aiming to mitigate intervention risks, activist funds seeking optimal investments, and investors looking to leverage potential stock price gains. Using data from the Russell 3000 index from 2016 to 2022, we evaluated 123 model configurations incorporating diverse imputation, oversampling, and machine learning techniques. Our best model achieved an AUC-ROC of 0.782, demonstrating its capability to effectively predict activist fund targets. To enhance interpretability, we employed the Shapley value method to identify key factors influencing a company’s likelihood of being targeted, highlighting the dynamic mechanisms underlying activist fund target selection. These insights offer a powerful tool for proactive corporate governance and informed investment strategies, advancing understanding of the mechanisms driving activist investment decisions.</div></div>","PeriodicalId":36340,"journal":{"name":"Journal of Finance and Data Science","volume":"10 ","pages":"Article 100146"},"PeriodicalIF":0.0,"publicationDate":"2024-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142746161","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-12-01DOI: 10.1016/j.jfds.2024.100145
Markus Leippold , Qian Wang , Min Yang
This paper explores the effectiveness of technical patterns in predicting asset prices and market movements, emphasizing the role of news sentiment. We employ an image recognition method to detect technical patterns in price images and assess whether this approach provides more information than traditional rule-based methods. Our findings indicate that many model-based patterns yield significant returns in the US market, whereas top-type patterns are less effective in the Chinese market. The model demonstrates high accuracy in training samples and strong out-of-sample performance. Our empirical analysis concludes that technical patterns remain effective in recent stock markets when combined with news sentiment, offering a profitable portfolio strategy. Moreover, we find patterns better predict returns for firms with high momentum, institutional ownership, and prior patterns in US, while in China, they are more effective for small firms with high momentum and institutional ownership. This study highlights the potential of image recognition methods in market data analysis and underscores the importance of sentiment in technical analysis.
{"title":"Technical patterns and news sentiment in stock markets","authors":"Markus Leippold , Qian Wang , Min Yang","doi":"10.1016/j.jfds.2024.100145","DOIUrl":"10.1016/j.jfds.2024.100145","url":null,"abstract":"<div><div>This paper explores the effectiveness of technical patterns in predicting asset prices and market movements, emphasizing the role of news sentiment. We employ an image recognition method to detect technical patterns in price images and assess whether this approach provides more information than traditional rule-based methods. Our findings indicate that many model-based patterns yield significant returns in the US market, whereas top-type patterns are less effective in the Chinese market. The model demonstrates high accuracy in training samples and strong out-of-sample performance. Our empirical analysis concludes that technical patterns remain effective in recent stock markets when combined with news sentiment, offering a profitable portfolio strategy. Moreover, we find patterns better predict returns for firms with high momentum, institutional ownership, and prior patterns in US, while in China, they are more effective for small firms with high momentum and institutional ownership. This study highlights the potential of image recognition methods in market data analysis and underscores the importance of sentiment in technical analysis.</div></div>","PeriodicalId":36340,"journal":{"name":"Journal of Finance and Data Science","volume":"10 ","pages":"Article 100145"},"PeriodicalIF":0.0,"publicationDate":"2024-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142746219","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-12-01DOI: 10.1016/j.jfds.2024.100147
Marzhan Alenova, Assem Utaliyeva, Ki-Joune Li
Hawala, a traditional but informal money transfer system, has been prevalent in many parts of the world, such as money laundering. Despite the regulatory actions taken by financial institutions, Hawala is still a key node in terror financing schemes and its extent of misuse is unknown. Due to the hidden transactions and limited knowledge about the Hawala, it is difficult for legal enforcement authorities such as financial intelligence units (FIU) of each country to detect and investigate the Hawala network. In this paper, we present a novel approach to detect the potential Hawala instances in the stream of financial transaction data by using graph mining techniques. In order to reflect the properties of Hawala, we apply graph mining methods such as graph centrality, Blackhole metric, and Hidden link metric as well as anomaly detection methods using graph convolutional network. Experiments demonstrate that the proposed method gives a meaningful result in detecting Hawala network and can be used as a complementary tool to the existing transactional monitoring tracks.
{"title":"Detecting Hawala network for money laundering by graph mining","authors":"Marzhan Alenova, Assem Utaliyeva, Ki-Joune Li","doi":"10.1016/j.jfds.2024.100147","DOIUrl":"10.1016/j.jfds.2024.100147","url":null,"abstract":"<div><div>Hawala, a traditional but informal money transfer system, has been prevalent in many parts of the world, such as money laundering. Despite the regulatory actions taken by financial institutions, Hawala is still a key node in terror financing schemes and its extent of misuse is unknown. Due to the hidden transactions and limited knowledge about the Hawala, it is difficult for legal enforcement authorities such as financial intelligence units (FIU) of each country to detect and investigate the Hawala network. In this paper, we present a novel approach to detect the potential Hawala instances in the stream of financial transaction data by using graph mining techniques. In order to reflect the properties of Hawala, we apply graph mining methods such as graph centrality, Blackhole metric, and Hidden link metric as well as anomaly detection methods using graph convolutional network. Experiments demonstrate that the proposed method gives a meaningful result in detecting Hawala network and can be used as a complementary tool to the existing transactional monitoring tracks.</div></div>","PeriodicalId":36340,"journal":{"name":"Journal of Finance and Data Science","volume":"10 ","pages":"Article 100147"},"PeriodicalIF":0.0,"publicationDate":"2024-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143175392","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-12-01DOI: 10.1016/j.jfds.2024.100148
Mustafa Pala, Emre Sefer
Non-fungible tokens (NFTs) are unique digital assets whose possession is defined over a blockchain. NFTs can represent multiple distinct objects such as art, images, videos, etc. There was a recent surge of interest in trading them which makes them another type of alternative investment. The inherent volatility of NFT prices, attributed to factors such as over-speculation, liquidity constraints, rarity, and market volatility, presents challenges for accurate price predictions. For such analysis and forecasting, machine learning methods offer a robust solution framework.
Here, we focus on three related prediction problems over NFTs: Predicting NFTs sale price, inferring whether a given NFT will participate in a secondary sale, and predicting NFT's sale price change over time. We analyze and learn the visual characteristics of NFTs by deep pre-trained models and combine such visual knowledge with additional important non-visual attributes such as the sale history, seller's and buyer's centralities in the trading network, and collection's resale probability. We categorize input NFTs into six categories based on their characteristics. Across detailed experiments, we found visual attributes obtained from deep pre-trained models to increase the prediction performance in all cases, and EfficientNet seems to perform the best. In general, CNN and XGBoost consistently outperformed the rest of them across all categories. We also publish our novel NFT dataset with temporal price knowledge, which is the first dataset to have NFT prices over time rather than at a single time point. Our code and NFT datasets are publicly available at https://github.com/seferlab/deep_nft.
{"title":"NFT price and sales characteristics prediction by transfer learning of visual attributes","authors":"Mustafa Pala, Emre Sefer","doi":"10.1016/j.jfds.2024.100148","DOIUrl":"10.1016/j.jfds.2024.100148","url":null,"abstract":"<div><div>Non-fungible tokens (NFTs) are unique digital assets whose possession is defined over a blockchain. NFTs can represent multiple distinct objects such as art, images, videos, etc. There was a recent surge of interest in trading them which makes them another type of alternative investment. The inherent volatility of NFT prices, attributed to factors such as over-speculation, liquidity constraints, rarity, and market volatility, presents challenges for accurate price predictions. For such analysis and forecasting, machine learning methods offer a robust solution framework.</div><div>Here, we focus on three related prediction problems over NFTs: Predicting NFTs sale price, inferring whether a given NFT will participate in a secondary sale, and predicting NFT's sale price change over time. We analyze and learn the visual characteristics of NFTs by deep pre-trained models and combine such visual knowledge with additional important non-visual attributes such as the sale history, seller's and buyer's centralities in the trading network, and collection's resale probability. We categorize input NFTs into six categories based on their characteristics. Across detailed experiments, we found visual attributes obtained from deep pre-trained models to increase the prediction performance in all cases, and EfficientNet seems to perform the best. In general, CNN and XGBoost consistently outperformed the rest of them across all categories. We also publish our novel NFT dataset with temporal price knowledge, which is the first dataset to have NFT prices over time rather than at a single time point. Our code and NFT datasets are publicly available at <span><span>https://github.com/seferlab/deep_nft</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":36340,"journal":{"name":"Journal of Finance and Data Science","volume":"10 ","pages":"Article 100148"},"PeriodicalIF":0.0,"publicationDate":"2024-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143175785","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-12-01DOI: 10.1016/j.jfds.2024.100143
Hugo Gobato Souto , Storm Koert Heuvel , Francisco Louzada Neto
This study evaluates the effectiveness of the TSMixer neural network model in forecasting stock realized volatility, comparing it with traditional and contemporary benchmark models. Using data from S&P 100 index stocks and three other datasets containing various financial securities, extensive analyses, including robustness tests, were conducted. Results show that TSMixer outperforms benchmark models in predicting individual stock volatility when applied to datasets with a large number of securities, leveraging its feature-mixing MLP techniques, which can properly model the financial tail dependence phenomenon. However, its superiority diminishes in datasets with fewer securities, such as stock indexes, foreign exchange rates, and commodities, where models like NBEATSx and NHITS often perform better. This indicates that TSMixer's performance is context-dependent, excelling when feature interdependencies can be fully exploited. The findings suggest that simplified neural network architectures like TSMixer can enhance forecasting accuracy in appropriate contexts but may have limitations in datasets with fewer securities.
{"title":"Time-mixing and feature-mixing modelling for realized volatility forecast: Evidence from TSMixer model","authors":"Hugo Gobato Souto , Storm Koert Heuvel , Francisco Louzada Neto","doi":"10.1016/j.jfds.2024.100143","DOIUrl":"10.1016/j.jfds.2024.100143","url":null,"abstract":"<div><div>This study evaluates the effectiveness of the TSMixer neural network model in forecasting stock realized volatility, comparing it with traditional and contemporary benchmark models. Using data from S&P 100 index stocks and three other datasets containing various financial securities, extensive analyses, including robustness tests, were conducted. Results show that TSMixer outperforms benchmark models in predicting individual stock volatility when applied to datasets with a large number of securities, leveraging its feature-mixing MLP techniques, which can properly model the financial tail dependence phenomenon. However, its superiority diminishes in datasets with fewer securities, such as stock indexes, foreign exchange rates, and commodities, where models like NBEATSx and NHITS often perform better. This indicates that TSMixer's performance is context-dependent, excelling when feature interdependencies can be fully exploited. The findings suggest that simplified neural network architectures like TSMixer can enhance forecasting accuracy in appropriate contexts but may have limitations in datasets with fewer securities.</div></div>","PeriodicalId":36340,"journal":{"name":"Journal of Finance and Data Science","volume":"10 ","pages":"Article 100143"},"PeriodicalIF":0.0,"publicationDate":"2024-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143175786","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-22DOI: 10.1016/j.jfds.2024.100142
Tingyu Qu
In financial markets, tail risks, representing the potential for substantial losses, bear significant implications for the formulation of effective risk management strategies. Yet, there exists a notable gap in understanding the interconnectedness within the global market, particularly when analysing time-series tail data. This study introduces a reliable method for identifying events indicative of tail transitions in financial time-series data. The investigation suggests consistent patterns governing extreme events across diverse industries and different time periods, suggestive of the financial contagion in tail risks. Importantly, time-series tail slopes in specific stocks emerge as viable predictors of price fluctuations in others. These findings offer valuable insights for portfolio diversification and risk mitigation in the interconnected financial market.
{"title":"Tail-driven portfolios: Unveiling financial contagion and enhancing risk management","authors":"Tingyu Qu","doi":"10.1016/j.jfds.2024.100142","DOIUrl":"10.1016/j.jfds.2024.100142","url":null,"abstract":"<div><div>In financial markets, tail risks, representing the potential for substantial losses, bear significant implications for the formulation of effective risk management strategies. Yet, there exists a notable gap in understanding the interconnectedness within the global market, particularly when analysing time-series tail data. This study introduces a reliable method for identifying events indicative of tail transitions in financial time-series data. The investigation suggests consistent patterns governing extreme events across diverse industries and different time periods, suggestive of the financial contagion in tail risks. Importantly, time-series tail slopes in specific stocks emerge as viable predictors of price fluctuations in others. These findings offer valuable insights for portfolio diversification and risk mitigation in the interconnected financial market.</div></div>","PeriodicalId":36340,"journal":{"name":"Journal of Finance and Data Science","volume":"10 ","pages":"Article 100142"},"PeriodicalIF":0.0,"publicationDate":"2024-10-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142659614","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-22DOI: 10.1016/j.jfds.2024.100144
Mikhail Stolbov , Maria Shchepeleva
We identify robust predictors of global systemic risk proxied by conditional capital shortfall (SRISK) among a comprehensive set of commodity prices for the period between January 2004 and December 2021. The search is based on a battery of ML variable selection algorithms which apply both to price levels and price shocks in the presence of control variables, including the first lag of SRISK, world industrial production, global economic policy uncertainty, geopolitical risk as well as the global stance of monetary and macroprudential policies. We find that these controls outweigh commodity prices as the predictors of global systemic risk. Of the commodities themselves, the prices for agricultural commodities, including food, e.g. chicken, bananas, beef, tea, cocoa, are more important predictors of global systemic risk than the prices for energy commodities, e.g. natural gas and oil prices. The financialization of agricultural commodities, bio-energy expansion as well as commodity-specific dependence of the major economies contributing to global systemic risk, e.g. China, account for our main finding. We also document the positive linkage between commodity prices and systemic risk for the majority of commodities. Thus, monitoring commodity prices to avoid their unbalanced growth is of vast importance to curb global systemic financial risk.
{"title":"Do commodity prices matter for global systemic risk? Evidence from ML variable selection","authors":"Mikhail Stolbov , Maria Shchepeleva","doi":"10.1016/j.jfds.2024.100144","DOIUrl":"10.1016/j.jfds.2024.100144","url":null,"abstract":"<div><div>We identify robust predictors of global systemic risk proxied by conditional capital shortfall (SRISK) among a comprehensive set of commodity prices for the period between January 2004 and December 2021. The search is based on a battery of ML variable selection algorithms which apply both to price levels and price shocks in the presence of control variables, including the first lag of SRISK, world industrial production, global economic policy uncertainty, geopolitical risk as well as the global stance of monetary and macroprudential policies. We find that these controls outweigh commodity prices as the predictors of global systemic risk. Of the commodities themselves, the prices for agricultural commodities, including food, e.g. chicken, bananas, beef, tea, cocoa, are more important predictors of global systemic risk than the prices for energy commodities, e.g. natural gas and oil prices. The financialization of agricultural commodities, bio-energy expansion as well as commodity-specific dependence of the major economies contributing to global systemic risk, e.g. China, account for our main finding. We also document the positive linkage between commodity prices and systemic risk for the majority of commodities. Thus, monitoring commodity prices to avoid their unbalanced growth is of vast importance to curb global systemic financial risk.</div></div>","PeriodicalId":36340,"journal":{"name":"Journal of Finance and Data Science","volume":"10 ","pages":"Article 100144"},"PeriodicalIF":0.0,"publicationDate":"2024-10-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142571479","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号