We employ a Least Absolute Shrinkage and Selection Operator (LASSO)-based extension of the Fama–MacBeth procedure to characterize the time-varying dependence of sovereign Credit Default Swap (CDS) spreads on macro indicators during the samples 2009–2013 and 2013–2020. While CDS spreads are mainly reflective of fundamentals, this relationship varies substantially over time, leading to price variation that appears unrelated to fundamentals. The estimated LASSO coefficients are used to endogenously identify macro-sensitivity “regimes” of variation, consistently with a multiple-equilibrium view of the sovereign debt markets.
{"title":"Anatomy of a Sovereign Debt Crisis: Machine Learning, Real-Time Macro Fundamentals, and CDS Spreads","authors":"Pierluigi Balduzzi, Roberto Savona, Lucia Alessi","doi":"10.1093/jjfinec/nbac021","DOIUrl":"https://doi.org/10.1093/jjfinec/nbac021","url":null,"abstract":"We employ a Least Absolute Shrinkage and Selection Operator (LASSO)-based extension of the Fama–MacBeth procedure to characterize the time-varying dependence of sovereign Credit Default Swap (CDS) spreads on macro indicators during the samples 2009–2013 and 2013–2020. While CDS spreads are mainly reflective of fundamentals, this relationship varies substantially over time, leading to price variation that appears unrelated to fundamentals. The estimated LASSO coefficients are used to endogenously identify macro-sensitivity “regimes” of variation, consistently with a multiple-equilibrium view of the sovereign debt markets.","PeriodicalId":47596,"journal":{"name":"Journal of Financial Econometrics","volume":"8 4","pages":""},"PeriodicalIF":2.5,"publicationDate":"2022-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138527908","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Risk premia are difficult to identify in nonstorable commodities such as electricity. In this article, we propose a modified Fama–French regression framework and show that when the spot prices do not follow a martingale—a common assumption in the electricity market—model specifications play an important role in detecting time-varying risk premia in the futures market. With this insight, we propose a multi-factor model that captures important dynamics in electricity prices and an estimation method based on particle Markov chain Monte Carlo to separate risk factors in energy prices. Using spot and futures data in the Germany/Austria electricity market, we demonstrate that our proposed model surpasses alternative models that ignore some of risk factors in forecasting spot prices and in detecting time-varying risk premia. Based on our proposed model, we separately identify risk premia carried by individual risk factors and document large variations in the premia of each factor.
{"title":"Identifying Risk Factors and Their Premia: A Study on Electricity Prices","authors":"Wei Wei, Asger Lunde","doi":"10.1093/jjfinec/nbac019","DOIUrl":"https://doi.org/10.1093/jjfinec/nbac019","url":null,"abstract":"\u0000 Risk premia are difficult to identify in nonstorable commodities such as electricity. In this article, we propose a modified Fama–French regression framework and show that when the spot prices do not follow a martingale—a common assumption in the electricity market—model specifications play an important role in detecting time-varying risk premia in the futures market. With this insight, we propose a multi-factor model that captures important dynamics in electricity prices and an estimation method based on particle Markov chain Monte Carlo to separate risk factors in energy prices. Using spot and futures data in the Germany/Austria electricity market, we demonstrate that our proposed model surpasses alternative models that ignore some of risk factors in forecasting spot prices and in detecting time-varying risk premia. Based on our proposed model, we separately identify risk premia carried by individual risk factors and document large variations in the premia of each factor.","PeriodicalId":47596,"journal":{"name":"Journal of Financial Econometrics","volume":" ","pages":""},"PeriodicalIF":2.5,"publicationDate":"2022-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49401792","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Kim Christensen, Mathias Siggaard, Bezirgen Veliyev
We inspect how accurate machine learning (ML) is at forecasting realized variance of the Dow Jones Industrial Average index constituents. We compare several ML algorithms, including regularization, regression trees, and neural networks, to multiple heterogeneous autoregressive (HAR) models. ML is implemented with minimal hyperparameter tuning. In spite of this, ML is competitive and beats the HAR lineage, even when the only predictors are the daily, weekly, and monthly lags of realized variance. The forecast gains are more pronounced at longer horizons. We attribute this to higher persistence in the ML models, which helps to approximate the long memory of realized variance. ML also excels at locating incremental information about future volatility from additional predictors. Lastly, we propose an ML measure of variable importance based on accumulated local effects. This shows that while there is agreement about the most important predictors, there is disagreement on their ranking, helping to reconcile our results.
{"title":"A Machine Learning Approach to Volatility Forecasting","authors":"Kim Christensen, Mathias Siggaard, Bezirgen Veliyev","doi":"10.1093/jjfinec/nbac020","DOIUrl":"https://doi.org/10.1093/jjfinec/nbac020","url":null,"abstract":"\u0000 We inspect how accurate machine learning (ML) is at forecasting realized variance of the Dow Jones Industrial Average index constituents. We compare several ML algorithms, including regularization, regression trees, and neural networks, to multiple heterogeneous autoregressive (HAR) models. ML is implemented with minimal hyperparameter tuning. In spite of this, ML is competitive and beats the HAR lineage, even when the only predictors are the daily, weekly, and monthly lags of realized variance. The forecast gains are more pronounced at longer horizons. We attribute this to higher persistence in the ML models, which helps to approximate the long memory of realized variance. ML also excels at locating incremental information about future volatility from additional predictors. Lastly, we propose an ML measure of variable importance based on accumulated local effects. This shows that while there is agreement about the most important predictors, there is disagreement on their ranking, helping to reconcile our results.","PeriodicalId":47596,"journal":{"name":"Journal of Financial Econometrics","volume":" ","pages":""},"PeriodicalIF":2.5,"publicationDate":"2022-06-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49533410","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper introduces new econometric tests to identify stochastic intensity jumps in high-frequency data. Our approach exploits the behavior of a time-varying stochastic intensity and allows us to assess how intensely stock market reacts to news. We describe the asymptotic properties of our test statistics, derive the associated central limit theorem and show in simulations that the tests have good size and reasonable power in finite-sample cases. Implementing our testing procedures on the S&P 500 exchange-traded fund data, we find strong evidence for the presence of intensity jumps surrounding the scheduled Federal Open Market Committee (FOMC) policy announcements. Intensity jumps occur very frequently, trigger sharp increases in realized volatility and arrive when differences in opinion among market participants are large at times of FOMC press releases. Unlike intensity jumps, volatility jumps fail to explain the variation in news-induced realized volatility.
{"title":"News Arrival, Time-Varying Jump Intensity, and Realized Volatility: Conditional Testing Approach","authors":"Deniz Erdemlioglu, Xiye Yang","doi":"10.1093/jjfinec/nbac015","DOIUrl":"https://doi.org/10.1093/jjfinec/nbac015","url":null,"abstract":"\u0000 This paper introduces new econometric tests to identify stochastic intensity jumps in high-frequency data. Our approach exploits the behavior of a time-varying stochastic intensity and allows us to assess how intensely stock market reacts to news. We describe the asymptotic properties of our test statistics, derive the associated central limit theorem and show in simulations that the tests have good size and reasonable power in finite-sample cases. Implementing our testing procedures on the S&P 500 exchange-traded fund data, we find strong evidence for the presence of intensity jumps surrounding the scheduled Federal Open Market Committee (FOMC) policy announcements. Intensity jumps occur very frequently, trigger sharp increases in realized volatility and arrive when differences in opinion among market participants are large at times of FOMC press releases. Unlike intensity jumps, volatility jumps fail to explain the variation in news-induced realized volatility.","PeriodicalId":47596,"journal":{"name":"Journal of Financial Econometrics","volume":" ","pages":""},"PeriodicalIF":2.5,"publicationDate":"2022-06-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43714323","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
We analyze the decomposition of the conditional, rather than the unconditional, variance of market returns based on an extension of the standard Campbell–Shiller approach. The relative importance of cash flow and discount rate news in determining the conditional variance of market returns exhibits significant variation over time and relates to economic conditions. The components of the conditional market variance outperform several benchmark variables, including the conditional market variance itself, in forecasting future market returns and realized variance across different horizons. The forecasts based on the conditional market variance components also provide sizable economic benefits compared with benchmark forecasts in an out-of-sample portfolio exercise where a myopic investor allocates her wealth between the market portfolio and a risk-free asset across different holding periods.
{"title":"Time Variation in Cash Flows and Discount Rates","authors":"Tolga Cenesizoglu, Denada Ibrushi","doi":"10.1093/jjfinec/nbac016","DOIUrl":"https://doi.org/10.1093/jjfinec/nbac016","url":null,"abstract":"We analyze the decomposition of the conditional, rather than the unconditional, variance of market returns based on an extension of the standard Campbell–Shiller approach. The relative importance of cash flow and discount rate news in determining the conditional variance of market returns exhibits significant variation over time and relates to economic conditions. The components of the conditional market variance outperform several benchmark variables, including the conditional market variance itself, in forecasting future market returns and realized variance across different horizons. The forecasts based on the conditional market variance components also provide sizable economic benefits compared with benchmark forecasts in an out-of-sample portfolio exercise where a myopic investor allocates her wealth between the market portfolio and a risk-free asset across different holding periods.","PeriodicalId":47596,"journal":{"name":"Journal of Financial Econometrics","volume":"10 3","pages":""},"PeriodicalIF":2.5,"publicationDate":"2022-06-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138527910","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Discussion of Identification Robust Testing of Risk Premia in Finite Samples","authors":"Francisco Peñaranda","doi":"10.1093/jjfinec/nbac014","DOIUrl":"https://doi.org/10.1093/jjfinec/nbac014","url":null,"abstract":"","PeriodicalId":47596,"journal":{"name":"Journal of Financial Econometrics","volume":" ","pages":""},"PeriodicalIF":2.5,"publicationDate":"2022-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44842558","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This study proposes a versatile model for the dynamics of the best bid and ask prices using an extended Hawkes process. The model incorporates the zero intensities of the spreadnarrowing processes at the minimum bid-ask spread, spread-dependent intensities, possible negative excitement, and nonnegative intensities. We apply the model to high-frequency best bid and ask price data from US stock markets. The empirical findings demonstrate a spread-narrowing tendency, excitations of the intensities caused by previous events, the impact of flash crashes, characteristic trends in fast trading over time, and the different features of market participants in the various exchanges.
{"title":"Modeling bid and ask price dynamics with an extended Hawkes process and its empirical applications for high-frequency stock market data","authors":"Kyungsub Lee, Byoung Ki Seo","doi":"10.1093/jjfinec/nbab029","DOIUrl":"https://doi.org/10.1093/jjfinec/nbab029","url":null,"abstract":"This study proposes a versatile model for the dynamics of the best bid and ask prices using an extended Hawkes process. The model incorporates the zero intensities of the spreadnarrowing processes at the minimum bid-ask spread, spread-dependent intensities, possible negative excitement, and nonnegative intensities. We apply the model to high-frequency best bid and ask price data from US stock markets. The empirical findings demonstrate a spread-narrowing tendency, excitations of the intensities caused by previous events, the impact of flash crashes, characteristic trends in fast trading over time, and the different features of market participants in the various exchanges.","PeriodicalId":47596,"journal":{"name":"Journal of Financial Econometrics","volume":"1 1","pages":""},"PeriodicalIF":2.5,"publicationDate":"2022-01-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"61098002","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：