Sjur Westgaard, Petter Osmundsen, Daniel Stenslet, J. Ringheim
A common perception in the literature is that oil price dynamics are most adequately explained by fundamental supply-and-demand factors. We use a general-to-specific approach and find that financial indicators are even more significant at modeling and predicting oil prices. We demonstrate empirically that the futures spreads level, high-yield bond spreads and PHLX Oil Service Sector (OSX) index are the best predictors of oil prices in the period February 2000–June 2013. (The OSX index is designed to track the performance of a set of companies involved in the oil services sector.) The OSX index is particularly interesting, as no study has analyzed its predictive power prior to our analysis. The relationship is intuitively meaningful, as stock prices, which strongly depend on the oil price, are determined in a market with well-informed investors that have strong incentives to gather correct market information. Moreover, the share prices serve as strong proxies or price signals, as they reflect future oil price expectations at any point of time. Furthermore, we demonstrate through an out-of-sample analysis that our most parsimonious model is superior to relevant benchmarks at forecasting oil price changes (two benchmarks were used: (1) a random walk and (2) ARIMA (2, 0, 2), which was optimized in-sample by minimizing the Akaike information criterion). Our findings do not necessarily imply that the financial sector determines oil prices. On the contrary, we take the view that fundamental information is traceable from financial markets, and, hence, financial predictors serve as indicators for oil price fundamentals.
{"title":"Modeling Superior Predictors for Crude Oil Prices","authors":"Sjur Westgaard, Petter Osmundsen, Daniel Stenslet, J. Ringheim","doi":"10.21314/JEM.2017.162","DOIUrl":"https://doi.org/10.21314/JEM.2017.162","url":null,"abstract":"A common perception in the literature is that oil price dynamics are most adequately explained by fundamental supply-and-demand factors. We use a general-to-specific approach and find that financial indicators are even more significant at modeling and predicting oil prices. We demonstrate empirically that the futures spreads level, high-yield bond spreads and PHLX Oil Service Sector (OSX) index are the best predictors of oil prices in the period February 2000–June 2013. (The OSX index is designed to track the performance of a set of companies involved in the oil services sector.) The OSX index is particularly interesting, as no study has analyzed its predictive power prior to our analysis. The relationship is intuitively meaningful, as stock prices, which strongly depend on the oil price, are determined in a market with well-informed investors that have strong incentives to gather correct market information. Moreover, the share prices serve as strong proxies or price signals, as they reflect future oil price expectations at any point of time. Furthermore, we demonstrate through an out-of-sample analysis that our most parsimonious model is superior to relevant benchmarks at forecasting oil price changes (two benchmarks were used: (1) a random walk and (2) ARIMA (2, 0, 2), which was optimized in-sample by minimizing the Akaike information criterion). Our findings do not necessarily imply that the financial sector determines oil prices. On the contrary, we take the view that fundamental information is traceable from financial markets, and, hence, financial predictors serve as indicators for oil price fundamentals.","PeriodicalId":445951,"journal":{"name":"ERN: Forecasting & Simulation (Prices) (Topic)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-06-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132493537","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper describes the econometric models used by the Banco de Espana to monitor consumer price inflation and forecast its future trends. The strategy followed heavily relies on the results from a set of econometric models, supplemented by expert judgment. We consider three different types of approaches and highlight the relevance of heterogeneity in price-setting behaviour and the importance of using models that allow for a slowly evolving local mean when forecasting inflation.
{"title":"A Suite of Inflation Forecasting Models","authors":"L. J. Álvarez, I. Sánchez-García","doi":"10.2139/ssrn.2924396","DOIUrl":"https://doi.org/10.2139/ssrn.2924396","url":null,"abstract":"This paper describes the econometric models used by the Banco de Espana to monitor consumer price inflation and forecast its future trends. The strategy followed heavily relies on the results from a set of econometric models, supplemented by expert judgment. We consider three different types of approaches and highlight the relevance of heterogeneity in price-setting behaviour and the importance of using models that allow for a slowly evolving local mean when forecasting inflation.","PeriodicalId":445951,"journal":{"name":"ERN: Forecasting & Simulation (Prices) (Topic)","volume":"144 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123480651","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In applied forecasting, there is a trade-off between in-sample fit and out-of-sample forecast accuracy. Parsimonious model specifications typically outperform richer model specifications. Consequently, there is often predictable information in forecast errors that is difficult to exploit. However, we show how this predictable information can be exploited in forecast combinations. In this case, optimal combination weights should minimize conditional mean squared error, or a conditional loss function, rather than the unconditional variance as in the commonly used framework of Bates and Granger (1969). We prove that our conditionally optimal weights lead to better forecast performance. The conditionally optimal weights support other forward-looking approaches to combining forecasts, where the forecast weights depend on the expected model performance. We show that forward-looking
{"title":"Conditionally Optimal Weights and Forward-Looking Approaches to Combining Forecasts","authors":"Christopher G. Gibbs, A. Vasnev","doi":"10.2139/ssrn.2919117","DOIUrl":"https://doi.org/10.2139/ssrn.2919117","url":null,"abstract":"In applied forecasting, there is a trade-off between in-sample fit and out-of-sample forecast accuracy. Parsimonious model specifications typically outperform richer model specifications. Consequently, there is often predictable information in forecast errors that is difficult to exploit. However, we show how this predictable information can be exploited in forecast combinations. In this case, optimal combination weights should minimize conditional mean squared error, or a conditional loss function, rather than the unconditional variance as in the commonly used framework of Bates and Granger (1969). We prove that our conditionally optimal weights lead to better forecast performance. The conditionally optimal weights support other forward-looking approaches to combining forecasts, where the forecast weights depend on the expected model performance. We show that forward-looking","PeriodicalId":445951,"journal":{"name":"ERN: Forecasting & Simulation (Prices) (Topic)","volume":"54 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-02-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134549044","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this paper, I study the nonparametric identification of a model of price discrimination with multidimensional consumer heterogeneity from disaggregated data on consumers' choices and characteristics. In particular, I consider the screening problem faced studied by Rochet and Chone (1998) where a seller of a product with multiple (and continuous) characteristics who only knows the joint density of consumer 'taste' and the production cost and chooses a product 'line' -- endogenous product characteristics. I determine the data features and additional conditions that are sufficient to identify the joint density of consumer heterogeneity, the cost function, and the utility functions that are common across consumers. If the product characteristics enter the utility function linearly, data from only one market is enough for identification, but if they enter nonlinearly we need data from at least two markets, or over two periods, with exogenous differences in costs. I also derive all testable restrictions imposed by the model on the data, i.e., the empirical content of the model, and also explore identification when prices are mismeasured and a product characteristic is missing.
{"title":"Identifying a Model of Screening with Multidimensional Consumer Heterogeneity","authors":"Gaurab Aryal","doi":"10.2139/ssrn.2531188","DOIUrl":"https://doi.org/10.2139/ssrn.2531188","url":null,"abstract":"In this paper, I study the nonparametric identification of a model of price discrimination with multidimensional consumer heterogeneity from disaggregated data on consumers' choices and characteristics. In particular, I consider the screening problem faced studied by Rochet and Chone (1998) where a seller of a product with multiple (and continuous) characteristics who only knows the joint density of consumer 'taste' and the production cost and chooses a product 'line' -- endogenous product characteristics. I determine the data features and additional conditions that are sufficient to identify the joint density of consumer heterogeneity, the cost function, and the utility functions that are common across consumers. If the product characteristics enter the utility function linearly, data from only one market is enough for identification, but if they enter nonlinearly we need data from at least two markets, or over two periods, with exogenous differences in costs. I also derive all testable restrictions imposed by the model on the data, i.e., the empirical content of the model, and also explore identification when prices are mismeasured and a product characteristic is missing.","PeriodicalId":445951,"journal":{"name":"ERN: Forecasting & Simulation (Prices) (Topic)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125662851","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
James A. Conover, David A. Dubofsky, Marilyn K. Wiley
Over the period 1970-2015, investment returns were enhanced by merely knowing concurrently whether the economy was in a state of expansion or contraction, and making the most basic asset allocation decision of whether to be in stocks or bonds. In the United States, an annual excess return of 2.01% was earned by investing in stocks during expansions and in bonds during contractions. In eight foreign markets, the average annual excess return from the same strategy was 1.74%. Forecasting business cycle troughs is more important than business cycle peaks. The authors conclude simple passive timing improves fund performance using business cycle peaks/troughs, and even slight forecasting prowess is rewarded with positive performance. Importantly, even investors who invested one month after the cycle turns could still earn excess returns.
{"title":"Does it Pay to Forecast the Business Cycle? A U.S. Update and an International Perspective","authors":"James A. Conover, David A. Dubofsky, Marilyn K. Wiley","doi":"10.2139/ssrn.2979464","DOIUrl":"https://doi.org/10.2139/ssrn.2979464","url":null,"abstract":"Over the period 1970-2015, investment returns were enhanced by merely knowing concurrently whether the economy was in a state of expansion or contraction, and making the most basic asset allocation decision of whether to be in stocks or bonds. In the United States, an annual excess return of 2.01% was earned by investing in stocks during expansions and in bonds during contractions. In eight foreign markets, the average annual excess return from the same strategy was 1.74%. Forecasting business cycle troughs is more important than business cycle peaks. The authors conclude simple passive timing improves fund performance using business cycle peaks/troughs, and even slight forecasting prowess is rewarded with positive performance. Importantly, even investors who invested one month after the cycle turns could still earn excess returns.","PeriodicalId":445951,"journal":{"name":"ERN: Forecasting & Simulation (Prices) (Topic)","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122901563","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This comment provides a correction to the paper "Is the Phillips Curve Alive and Well After All? Inflation Expectations and the Missing Disinflation" by Olivier Coibion and Yuriy Gorodnichenko (2015) in the American Economic Journal: Macroeconomics.
{"title":"Comment: 'Is the Phillips Curve Alive and Well after All? Inflation Expectations and the Missing Disinflation'","authors":"C. Binder","doi":"10.2139/ssrn.2789901","DOIUrl":"https://doi.org/10.2139/ssrn.2789901","url":null,"abstract":"This comment provides a correction to the paper \"Is the Phillips Curve Alive and Well After All? Inflation Expectations and the Missing Disinflation\" by Olivier Coibion and Yuriy Gorodnichenko (2015) in the American Economic Journal: Macroeconomics.","PeriodicalId":445951,"journal":{"name":"ERN: Forecasting & Simulation (Prices) (Topic)","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116566405","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
We assess to what extent indicators of financial conditions can be considered relevant determinants and predictors of macroeconomic aggregates. The main finding is that controlling for default risk and risk aversion measures improves the forecasts of output, employment and loans, but that this improvement is largely attributable to the recession periods of 2001 and 2008. A structural VAR analysis further reveals that financial condition indicators display significant real effects only after the Great Financial Crisis. In particular, an unexpected increase in the credit spread in 2010 causes an output contraction that lasts for about two years, with an annualised through of 4.8%, and explains up to 35% of the forecast error variance of industrial production.
{"title":"Macroeconomic Activity and Risk Indicators: An Unstable Relationship","authors":"Angela Abbate, Massimiliano Marcellino","doi":"10.2139/ssrn.2980643","DOIUrl":"https://doi.org/10.2139/ssrn.2980643","url":null,"abstract":"We assess to what extent indicators of financial conditions can be considered relevant determinants and predictors of macroeconomic aggregates. The main finding is that controlling for default risk and risk aversion measures improves the forecasts of output, employment and loans, but that this improvement is largely attributable to the recession periods of 2001 and 2008. A structural VAR analysis further reveals that financial condition indicators display significant real effects only after the Great Financial Crisis. In particular, an unexpected increase in the credit spread in 2010 causes an output contraction that lasts for about two years, with an annualised through of 4.8%, and explains up to 35% of the forecast error variance of industrial production.","PeriodicalId":445951,"journal":{"name":"ERN: Forecasting & Simulation (Prices) (Topic)","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-05-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116794432","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
W. Bruine de Bruin, Wilbert van der Klaauw, M. van Rooij, Federica Teppa, Klaas de Vos
Several national surveys aim to elicit consumers’ inflation expectations. Median expectations tend to track objective inflation estimates over time, although responses display large dispersion. Medians also tend to differ between surveys, possibly reflecting survey design differences. Using a nationally representative Dutch sample, we evaluate the importance of three survey design features in explaining observed differences: mode (face-to-face vs. web), question wording (‘prices in general’ vs. ‘inflation’), and the explicit opportunity to revise responses. We examine effects on item non-responses, revisions, reported inflation expectations and their deviation from the CPI inflation rate. We discuss implications of our findings for survey design.
{"title":"Measuring Expectations of Inflation: Effects of Survey Mode, Wording, and Opportunities to Revise","authors":"W. Bruine de Bruin, Wilbert van der Klaauw, M. van Rooij, Federica Teppa, Klaas de Vos","doi":"10.2139/ssrn.2745831","DOIUrl":"https://doi.org/10.2139/ssrn.2745831","url":null,"abstract":"Several national surveys aim to elicit consumers’ inflation expectations. Median expectations tend to track objective inflation estimates over time, although responses display large dispersion. Medians also tend to differ between surveys, possibly reflecting survey design differences. Using a nationally representative Dutch sample, we evaluate the importance of three survey design features in explaining observed differences: mode (face-to-face vs. web), question wording (‘prices in general’ vs. ‘inflation’), and the explicit opportunity to revise responses. We examine effects on item non-responses, revisions, reported inflation expectations and their deviation from the CPI inflation rate. We discuss implications of our findings for survey design.","PeriodicalId":445951,"journal":{"name":"ERN: Forecasting & Simulation (Prices) (Topic)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128050069","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
We construct a text-based measure of uncertainty starting in 1890 using front-page articles of the Wall Street Journal. News implied volatility (NVIX) peaks during stock market crashes, times of policy-related uncertainty, world wars, and financial crises. In US postwar data, periods when NVIX is high are followed by periods of above average stock returns, even after controlling for contemporaneous and forward-looking measures of stock market volatility. News coverage related to wars and government policy explains most of the time variation in risk premia our measure identifies. Over the longer 1890–2009 sample that includes the Great Depression and two world wars, high NVIX predicts high future returns in normal times and rises just before transitions into economic disasters. The evidence is consistent with recent theories emphasizing time variation in rare disaster risk as a source of aggregate asset prices fluctuations.
{"title":"News Implied Volatility and Disaster Concerns","authors":"Asaf Manela, Alan Moreira","doi":"10.2139/ssrn.2382197","DOIUrl":"https://doi.org/10.2139/ssrn.2382197","url":null,"abstract":"We construct a text-based measure of uncertainty starting in 1890 using front-page articles of the Wall Street Journal. News implied volatility (NVIX) peaks during stock market crashes, times of policy-related uncertainty, world wars, and financial crises. In US postwar data, periods when NVIX is high are followed by periods of above average stock returns, even after controlling for contemporaneous and forward-looking measures of stock market volatility. News coverage related to wars and government policy explains most of the time variation in risk premia our measure identifies. Over the longer 1890–2009 sample that includes the Great Depression and two world wars, high NVIX predicts high future returns in normal times and rises just before transitions into economic disasters. The evidence is consistent with recent theories emphasizing time variation in rare disaster risk as a source of aggregate asset prices fluctuations.","PeriodicalId":445951,"journal":{"name":"ERN: Forecasting & Simulation (Prices) (Topic)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131392356","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
We define and forecast classical business cycle turning points for the Norwegian economy. When defining reference business cycles, we compare a univariate and a multivariate Bry–Boschan approach with univariate Markov-switching models and Markov-switching factor models. On the basis of a receiver operating characteristic curve methodology and a comparison of the business cycle turning points of Norway’s main trading partners, we find that a Markov-switching factor model provides the most reasonable definition of Norwegian business cycles for the sample 1978Q1–2011Q4. In a real-time out-of-sample forecasting exercise, focusing on the last recession, we show that univariate Markov-switching models applied to surveys and a financial conditions index are timely and accurate in calling the last peak in real time. However, the models are less accurate and timely in calling the trough in real time.
{"title":"Identification and Real-Time Forecasting of Norwegian Business Cycles","authors":"K. Aastveit, A. Jore, F. Ravazzolo","doi":"10.2139/ssrn.2616800","DOIUrl":"https://doi.org/10.2139/ssrn.2616800","url":null,"abstract":"We define and forecast classical business cycle turning points for the Norwegian economy. When defining reference business cycles, we compare a univariate and a multivariate Bry–Boschan approach with univariate Markov-switching models and Markov-switching factor models. On the basis of a receiver operating characteristic curve methodology and a comparison of the business cycle turning points of Norway’s main trading partners, we find that a Markov-switching factor model provides the most reasonable definition of Norwegian business cycles for the sample 1978Q1–2011Q4. In a real-time out-of-sample forecasting exercise, focusing on the last recession, we show that univariate Markov-switching models applied to surveys and a financial conditions index are timely and accurate in calling the last peak in real time. However, the models are less accurate and timely in calling the trough in real time.","PeriodicalId":445951,"journal":{"name":"ERN: Forecasting & Simulation (Prices) (Topic)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-05-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121780122","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","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：