We show that the minimum description length (MDL) criterion widely used to estimate lin- ear change-point (CP) models corresponds to the marginal likelihood of a Bayesian model with a specific class of prior distributions. This allows for results from the frequentist and Bayesian literatures to be bridged together. In this estimation framework, one can rely on the consistency of the number and locations of the estimated CPs and the computational efficiency of frequentist methods, and obtain a probability of observing a CP at a given time, compute model posterior probabilities, and select or combine CP methods via Bayesian posteriors. This approach is further extended to other popular information criteria (such as Akaike, Bayes, and Hannan-Quinn’s). Moreover, we propose several CP methods that take advantage of the MDL probabilistic representation. Based on simulated and macroeconomic data, the novel methods detect and date structural breaks with the same or improved level of accuracy than state-of-the- art approaches. Finally, we highlight the usefulness of combining CP methods for long time series, both in terms of improved detection accuracy and reduced computational cost.
{"title":"Frequentist and Bayesian Change-Point Models: A Missing Link","authors":"David Ardia, A. Dufays, C. O. Criado","doi":"10.2139/ssrn.3499824","DOIUrl":"https://doi.org/10.2139/ssrn.3499824","url":null,"abstract":"We show that the minimum description length (MDL) criterion widely used to estimate lin- ear change-point (CP) models corresponds to the marginal likelihood of a Bayesian model with a specific class of prior distributions. This allows for results from the frequentist and Bayesian literatures to be bridged together. In this estimation framework, one can rely on the consistency of the number and locations of the estimated CPs and the computational efficiency of frequentist methods, and obtain a probability of observing a CP at a given time, compute model posterior probabilities, and select or combine CP methods via Bayesian posteriors. This approach is further extended to other popular information criteria (such as Akaike, Bayes, and Hannan-Quinn’s). Moreover, we propose several CP methods that take advantage of the MDL probabilistic representation. Based on simulated and macroeconomic data, the novel methods detect and date structural breaks with the same or improved level of accuracy than state-of-the- art approaches. Finally, we highlight the usefulness of combining CP methods for long time series, both in terms of improved detection accuracy and reduced computational cost.","PeriodicalId":308524,"journal":{"name":"ERN: Other Econometrics: Applied Econometric Modeling in Forecasting (Topic)","volume":"22 7","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120857326","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}
Change-point (CP) VAR models face a dimensionality curse due to the proliferation of parameters that arises when new breaks are detected. To handle large data sets, we introduce the Sparse CP-VAR model that determines which parameters truly vary when a break is detected. By doing so, the number of new parameters to estimate at each regime is drastically reduced and the CP dynamic becomes easier to interpret. The Sparse CP-VAR model disentangles the dynamics of the mean parameters and the covariance matrix. The former uses CP dynamics with shrinkage prior distributions while the latter is driven by an infinite hidden Markov framework. A simulation study highlights that the framework detects correctly the number of breaks per model parameter, and that it takes advantage of common breaks in the cross-sectional dimension to more precisely estimate them. Our applications on financial and macroeconomic systems highlight that the Sparse CP-VAR model helps interpreting the detected breaks. It turns out that many spillover effects have zero regimes meaning that they are zero for the entire sample period. Forecasting wise, the Sparse CP-VAR model is competitive against several recent time-varying parameter and CP-VAR models in terms of log predictive densities.
{"title":"Sparse Change-Point VAR models","authors":"A. Dufays, Li Zhuo, J. Rombouts, Yong Song","doi":"10.2139/ssrn.3461692","DOIUrl":"https://doi.org/10.2139/ssrn.3461692","url":null,"abstract":"Change-point (CP) VAR models face a dimensionality curse due to the proliferation of parameters that arises when new breaks are detected. To handle large data sets, we introduce the Sparse CP-VAR model that determines which parameters truly vary when a break is detected. By doing so, the number of new parameters to estimate at each regime is drastically reduced and the CP dynamic becomes easier to interpret. The Sparse CP-VAR model disentangles the dynamics of the mean parameters and the covariance matrix. The former uses CP dynamics with shrinkage prior distributions while the latter is driven by an infinite hidden Markov framework. A simulation study highlights that the framework detects correctly the number of breaks per model parameter, and that it takes advantage of common breaks in the cross-sectional dimension to more precisely estimate them. Our applications on financial and macroeconomic systems highlight that the Sparse CP-VAR model helps interpreting the detected breaks. It turns out that many spillover effects have zero regimes meaning that they are zero for the entire sample period. Forecasting wise, the Sparse CP-VAR model is competitive against several recent time-varying parameter and CP-VAR models in terms of log predictive densities.","PeriodicalId":308524,"journal":{"name":"ERN: Other Econometrics: Applied Econometric Modeling in Forecasting (Topic)","volume":"59 2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-08-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116521085","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}
Russian Abstract: Предложен комплекс опережающих индикаторов инновационной динамики России и их многокомпонентные оценки для 78 регионов за период 2005-2017 годов. Индикаторы характеризуют динамику основных отраслей российской экономики (строительства, торговли, добычи полезных ископаемых, обрабатывающей промышленности) трендами, циклическими, сезонными колебаниями и их взаимодействиями. Динамика продукции сельского хозяйства, в большей степени обусловленная климатическими и природными условиями, пока не рассматривались. Для трендов индикаторов предложены одна линейная и шесть существенно нелинейных (нелинейных по параметрам) моделей. Сезонная компонента моделировалась гармоникой с сезонными коэффициентами, а циклы Китчина, Жугляра и Кузнеца - суммой гармоник с некратными частотами (по Е.Е. Слуцкому). Взаимодействие компонент рассматривалось как линейное (аддитивной), так и нелинейное (аддитивно-мультипликативное). Скорректированный коэффициент детерминации обосновал более точные модели. Было уделено внимание расширению адаптации инструментария, прогнозированию всех регулярных компонент индикаторов, характеристикам инновационного развития и синхронности циклов отдельных регионов. Представлен новый и большей точности материал для руководителей, служб и предприятий регионов, определены дальнейшие перспективы развития предложенного инструментария.
English Abstract: The authors proposed a set of leading indicators of innovation dynamics in Russia and their multicomponent estimates for 78 regions during 2005-2017. The indicators show dynamics of the most important economic sectors in Russia (building, trade, mining, manufacturing and its branches) while agricultural production, which dynamics mostly depends on climate and geography, have not yet been considered The models include trends, cycles, seasonal component and their interactions. For trends one linear and six substantially nonlinear (nonlinear in the parameters) models are used. The seasonal component was modeled by seasonal coefficients. Kitchin, Juglar and Kuznets cycles was modeled using sum of three sine curves with non-proportional frequencies (as suggested by E.Slutsky). The interaction of components was considered both linear (additive) and nonlinear (additive-multiplicative). The most accurate models were justified using adjusted coefficient of determination. Special attention is paid to adaptive modeling tools expansion, leading indicators decomposition and forecasting, innovative development analysis and regional cycles synchrony or asynchrony. As a result of the modeling the authors presented new and more accurate material for regional authorities and managers. Further development of the proposed modeling tools are also suggested.
{"title":"Многомодельная Оценка Инновационного Развития 78 Российских Регионов По Опережающим Индикаторам За Период 2005–2017гг. (Multimodel Estimation for Innovative Development of 78 Russian Regions Using Leading Indicators During 2005-2017)","authors":"V. Semenychev, Anastasiya Korobetskaya","doi":"10.2139/ssrn.3373579","DOIUrl":"https://doi.org/10.2139/ssrn.3373579","url":null,"abstract":"<b>Russian Abstract:</b> Предложен комплекс опережающих индикаторов инновационной динамики России и их многокомпонентные оценки для 78 регионов за период 2005-2017 годов. Индикаторы характеризуют динамику основных отраслей российской экономики (строительства, торговли, добычи полезных ископаемых, обрабатывающей промышленности) трендами, циклическими, сезонными колебаниями и их взаимодействиями. Динамика продукции сельского хозяйства, в большей степени обусловленная климатическими и природными условиями, пока не рассматривались. Для трендов индикаторов предложены одна линейная и шесть существенно нелинейных (нелинейных по параметрам) моделей. Сезонная компонента моделировалась гармоникой с сезонными коэффициентами, а циклы Китчина, Жугляра и Кузнеца - суммой гармоник с некратными частотами (по Е.Е. Слуцкому). Взаимодействие компонент рассматривалось как линейное (аддитивной), так и нелинейное (аддитивно-мультипликативное). Скорректированный коэффициент детерминации обосновал более точные модели. Было уделено внимание расширению адаптации инструментария, прогнозированию всех регулярных компонент индикаторов, характеристикам инновационного развития и синхронности циклов отдельных регионов. Представлен новый и большей точности материал для руководителей, служб и предприятий регионов, определены дальнейшие перспективы развития предложенного инструментария.<br><br><b>English Abstract:</b> The authors proposed a set of leading indicators of innovation dynamics in Russia and their multicomponent estimates for 78 regions during 2005-2017. The indicators show dynamics of the most important economic sectors in Russia (building, trade, mining, manufacturing and its branches) while agricultural production, which dynamics mostly depends on climate and geography, have not yet been considered The models include trends, cycles, seasonal component and their interactions. For trends one linear and six substantially nonlinear (nonlinear in the parameters) models are used. The seasonal component was modeled by seasonal coefficients. Kitchin, Juglar and Kuznets cycles was modeled using sum of three sine curves with non-proportional frequencies (as suggested by E.Slutsky). The interaction of components was considered both linear (additive) and nonlinear (additive-multiplicative). The most accurate models were justified using adjusted coefficient of determination. Special attention is paid to adaptive modeling tools expansion, leading indicators decomposition and forecasting, innovative development analysis and regional cycles synchrony or asynchrony. As a result of the modeling the authors presented new and more accurate material for regional authorities and managers. Further development of the proposed modeling tools are also suggested.","PeriodicalId":308524,"journal":{"name":"ERN: Other Econometrics: Applied Econometric Modeling in Forecasting (Topic)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131703048","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}
R. Casarin, S. Grassi, Francesco Ravazzollo, H. V. Dijk
A flexible forecast density combination approach is introduced that can deal with large data sets. It extends the mixture of experts approach by allowing for model set incompleteness and dynamic learning of combination weights. A dimension reduction step is introduced using a sequential clustering mechanism that allocates the large set of forecast densities into a small number of subsets and the combination weights of the large set of densities are modelled as a dynamic factor model with a number of factors equal to the number of subsets. The forecast density combination is represented as a large finite mixture in nonlinear state space form. An efficient simulation-based Bayesian inferential procedure is proposed using parallel sequential clustering and filtering, implemented on graphics processing units. The approach is applied to track the Standard & Poor 500 index combining more than 7000 forecast densities based on 1856 US individual stocks that are are clustered in a relatively small subset. Substantial forecast and economic gains are obtained, in particular, in the tails using Value-at-Risk. Using a large macroeconomic data set of 142 series, similar forecast gains, including probabilities of recession, are obtained from multivariate forecast density combinations of US real GDP, Inflation, Treasury Bill yield and Employment. Evidence obtained on the dynamic patterns in the financial as well as macroeconomic clusters provide valuable signals useful for improved modelling and more effective economic and financial policies.
{"title":"Forecast Density Combinations with Dynamic Learning for Large Data Sets in Economics and Finance","authors":"R. Casarin, S. Grassi, Francesco Ravazzollo, H. V. Dijk","doi":"10.2139/ssrn.3363556","DOIUrl":"https://doi.org/10.2139/ssrn.3363556","url":null,"abstract":"A flexible forecast density combination approach is introduced that can deal with large data sets. It extends the mixture of experts approach by allowing for model set incompleteness and dynamic learning of combination weights. A dimension reduction step is introduced using a sequential clustering mechanism that allocates the large set of forecast densities into a small number of subsets and the combination weights of the large set of densities are modelled as a dynamic factor model with a number of factors equal to the number of subsets. The forecast density combination is represented as a large finite mixture in nonlinear state space form. An efficient simulation-based Bayesian inferential procedure is proposed using parallel sequential clustering and filtering, implemented on graphics processing units. The approach is applied to track the Standard & Poor 500 index combining more than 7000 forecast densities based on 1856 US individual stocks that are are clustered in a relatively small subset. Substantial forecast and economic gains are obtained, in particular, in the tails using Value-at-Risk. Using a large macroeconomic data set of 142 series, similar forecast gains, including probabilities of recession, are obtained from multivariate forecast density combinations of US real GDP, Inflation, Treasury Bill yield and Employment. Evidence obtained on the dynamic patterns in the financial as well as macroeconomic clusters provide valuable signals useful for improved modelling and more effective economic and financial policies.","PeriodicalId":308524,"journal":{"name":"ERN: Other Econometrics: Applied Econometric Modeling in Forecasting (Topic)","volume":"195 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-03-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121850367","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 study proposes a real-time estimate of inattention, based on micro-level data. I show that a simple specification that estimates the persistence of a forecaster's deviation from the mean provides a direct estimate of parameters of information frictions according to prominent models of expectations. The new estimate can also be interpreted as a hybrid measure of both information frictions and behavioral frictions. Using the new specification, I revise several key findings documented in the previous literature. I find higher levels of inattention and document new forms of variations over time and across variables, horizons, individuals, and types of agents. I also report new results from long-run forecasts and document an unprecedented response to COVID-19.
{"title":"Tracking Inattention","authors":"Nathan G. Goldstein","doi":"10.2139/ssrn.3641353","DOIUrl":"https://doi.org/10.2139/ssrn.3641353","url":null,"abstract":"\u0000 This study proposes a real-time estimate of inattention, based on micro-level data. I show that a simple specification that estimates the persistence of a forecaster's deviation from the mean provides a direct estimate of parameters of information frictions according to prominent models of expectations. The new estimate can also be interpreted as a hybrid measure of both information frictions and behavioral frictions. Using the new specification, I revise several key findings documented in the previous literature. I find higher levels of inattention and document new forms of variations over time and across variables, horizons, individuals, and types of agents. I also report new results from long-run forecasts and document an unprecedented response to COVID-19.","PeriodicalId":308524,"journal":{"name":"ERN: Other Econometrics: Applied Econometric Modeling in Forecasting (Topic)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-03-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129048107","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 study analyses point forecasts of exact scoreline outcomes for football matches in the English Premier League. These forecasts were made for distinct competitions and originally judged differently. We compare these with implied probability forecasts using bookmaker odds and a crowd of tipsters, as well as point and probability forecasts generated from a statistical model. From evaluating these sources and types of forecast, using various methods, we argue that regression encompassing is the most appropriate way to compare point and probability forecasts, and find that both these types of forecasts for football match scorelines generally add information to one another.
{"title":"Evaluating Strange Forecasts: The Curious Case of Football Match Scorelines","authors":"J. Reade, Carl Singleton, Alasdair Brown","doi":"10.2139/ssrn.3340598","DOIUrl":"https://doi.org/10.2139/ssrn.3340598","url":null,"abstract":"This study analyses point forecasts of exact scoreline outcomes for football matches in the English Premier League. These forecasts were made for distinct competitions and originally judged differently. We compare these with implied probability forecasts using bookmaker odds and a crowd of tipsters, as well as point and probability forecasts generated from a statistical model. From evaluating these sources and types of forecast, using various methods, we argue that regression encompassing is the most appropriate way to compare point and probability forecasts, and find that both these types of forecasts for football match scorelines generally add information to one another.","PeriodicalId":308524,"journal":{"name":"ERN: Other Econometrics: Applied Econometric Modeling in Forecasting (Topic)","volume":"29 1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120889941","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 employ averaging over statistical ensemble of assets to derive an index characterizing the level of correlations in a financial market – the eCORR index. This index does not require lengthy historical data and reacts immediately to any changes in correlations. Study of statistical properties of eCORR for US equity markets reveals how volatility is distributed between the common part and the part specific to individual equities. It also allows to demonstrate and quantify the correlation-drawdown hysteresis effect. The eCORR index promises to be useful for early detection of market correlations, managing risk concentrations and maintaining portfolio diversification.
{"title":"Measurement of Current Market Correlations Based on Ensemble Statistics","authors":"Jack Sarkissian, Joel H. Sebold","doi":"10.2139/ssrn.3308669","DOIUrl":"https://doi.org/10.2139/ssrn.3308669","url":null,"abstract":"We employ averaging over statistical ensemble of assets to derive an index characterizing the level of correlations in a financial market – the eCORR index. This index does not require lengthy historical data and reacts immediately to any changes in correlations. Study of statistical properties of eCORR for US equity markets reveals how volatility is distributed between the common part and the part specific to individual equities. It also allows to demonstrate and quantify the correlation-drawdown hysteresis effect. The eCORR index promises to be useful for early detection of market correlations, managing risk concentrations and maintaining portfolio diversification.","PeriodicalId":308524,"journal":{"name":"ERN: Other Econometrics: Applied Econometric Modeling in Forecasting (Topic)","volume":"75 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116064944","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}
Pub Date : 2018-08-01DOI: 10.5089/9781484372555.001
Svetlana Cerovic, Kerstin Gerling, Andrew Hodge, Paulo A. Medas
This paper identifies leading indicators of fiscal crises based on a large sample of countries at different stages of development over 1970-2015. Our results are robust to different methodologies and sample periods. Previous literature on early warning sistems (EWS) for fiscal crises is scarce and based on small samples of advanced and emerging markets, raising doubts about the robustness of the results. Using a larger sample, our analysis shows that both nonfiscal (external and internal imbalances) and fiscal variables help predict crises among advanced and emerging economies. Our models performed well in out-of-sample forecasting and in predicting the most recent crises, a weakness of EWS in general. We also build EWS for low income countries, which had been overlooked in the literature.
{"title":"Predicting Fiscal Crises","authors":"Svetlana Cerovic, Kerstin Gerling, Andrew Hodge, Paulo A. Medas","doi":"10.5089/9781484372555.001","DOIUrl":"https://doi.org/10.5089/9781484372555.001","url":null,"abstract":"This paper identifies leading indicators of fiscal crises based on a large sample of countries at different stages of development over 1970-2015. Our results are robust to different methodologies and sample periods. Previous literature on early warning sistems (EWS) for fiscal crises is scarce and based on small samples of advanced and emerging markets, raising doubts about the robustness of the results. Using a larger sample, our analysis shows that both nonfiscal (external and internal imbalances) and fiscal variables help predict crises among advanced and emerging economies. Our models performed well in out-of-sample forecasting and in predicting the most recent crises, a weakness of EWS in general. We also build EWS for low income countries, which had been overlooked in the literature.","PeriodicalId":308524,"journal":{"name":"ERN: Other Econometrics: Applied Econometric Modeling in Forecasting (Topic)","volume":"282 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131651270","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}
Using a large number of predictors and based on an extended iterated combination approach of Lin, Wu, and Zhou (2017), we document both statistical and economic significance of Treasury bond return predictability. Macroeconomic and aggregate liquidity variables contain predictive information for bond returns and combining them with term structure and Ludvigson-Ng macro factors significantly improve out-of-sample forecast gains. We also find that variance forecasts can be substantially improved with our approach, yielding significant gains in asset allocation decision. Our results show that information from a large number of predictors collectively contributes to the time-varying Treasury bond premia, and this is robust to different return measures, horizons and sample periods.
{"title":"Iterated Combination Forecast and Treasury Bond Predictability","authors":"Hai Lin, Wenjie Liu, Chunchi Wu, Guofu Zhou","doi":"10.2139/ssrn.3220751","DOIUrl":"https://doi.org/10.2139/ssrn.3220751","url":null,"abstract":"Using a large number of predictors and based on an extended iterated combination approach of Lin, Wu, and Zhou (2017), we document both statistical and economic significance of Treasury bond return predictability. Macroeconomic and aggregate liquidity variables contain predictive information for bond returns and combining them with term structure and Ludvigson-Ng macro factors significantly improve out-of-sample forecast gains. We also find that variance forecasts can be substantially improved with our approach, yielding significant gains in asset allocation decision. Our results show that information from a large number of predictors collectively contributes to the time-varying Treasury bond premia, and this is robust to different return measures, horizons and sample periods.","PeriodicalId":308524,"journal":{"name":"ERN: Other Econometrics: Applied Econometric Modeling in Forecasting (Topic)","volume":"434-435 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-07-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121469654","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 improve on our Vigilant Asset Allocation (VAA) by the introduction of a separate “canary” universe for signaling the need for crash protection, using the concept of breadth momentum. The amount of cash is now governed by the number of canary assets with bad (non-positive) momentum. The risky part is still based on relative momentum (or relative strength), just like VAA. We call this strategy Defensive Assets Allocation (DAA). The aim of DAA is to lower the average cash (or bond) fraction while keeping nearly the same degree of crash protection as with VAA. Using a very simple model from Dec 1926 to Dec 1970 with only the SP500 index as risky asset, we find an optimal canary universe of VWO and BND (aka EEM and AGG), which turns out to be rather effective also for nearly all our VAA universes, from Dec 1970 to Mar 2018. The average cash fraction of DAA is often less than half that of VAA’s, while return and risk are similar and for recent years even better. The usage of a separate “canary” universe for signaling the need for crash protection also improves the tracking error with respect to the passive (buy-and-hold) benchmark and limits turnover.
{"title":"Breadth Momentum and the Canary Universe: Defensive Asset Allocation (DAA)","authors":"W. Keller, Jan Willem Keuning","doi":"10.2139/ssrn.3212862","DOIUrl":"https://doi.org/10.2139/ssrn.3212862","url":null,"abstract":"We improve on our Vigilant Asset Allocation (VAA) by the introduction of a separate “canary” universe for signaling the need for crash protection, using the concept of breadth momentum. The amount of cash is now governed by the number of canary assets with bad (non-positive) momentum. The risky part is still based on relative momentum (or relative strength), just like VAA. We call this strategy Defensive Assets Allocation (DAA). The aim of DAA is to lower the average cash (or bond) fraction while keeping nearly the same degree of crash protection as with VAA. Using a very simple model from Dec 1926 to Dec 1970 with only the SP500 index as risky asset, we find an optimal canary universe of VWO and BND (aka EEM and AGG), which turns out to be rather effective also for nearly all our VAA universes, from Dec 1970 to Mar 2018. The average cash fraction of DAA is often less than half that of VAA’s, while return and risk are similar and for recent years even better. The usage of a separate “canary” universe for signaling the need for crash protection also improves the tracking error with respect to the passive (buy-and-hold) benchmark and limits turnover.","PeriodicalId":308524,"journal":{"name":"ERN: Other Econometrics: Applied Econometric Modeling in Forecasting (Topic)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131146550","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：