Portfolio managers attract asset owners by promoting their skills and investment decision process. Performance attribution explains how their decisions add value to portfolios. To be useful, the attribution framework must reflect the decisions made by the managers. Country and currency exposures of global portfolios are often managed with distinct strategies. The author demonstrates that the standard implementation of the Brinson–Fachler attribution framework does not reflect those strategies and produces non-intuitive results. To reflect those strategies, the author shows ways of implementing the Brinson–Fachler framework that incorporate the cost of hedging and the currency surprise and contrasts the results with the standard implementation. He highlights the similarities and the divergences between the Karnosky–Singer and Ankrim–Hensel approaches and proposes a technique to make both approaches equivalent.
{"title":"Multicurrency Performance Attribution Analysis with Currency Overlay Management","authors":"C. Giguere","doi":"10.3905/jpm.2023.1.516","DOIUrl":"https://doi.org/10.3905/jpm.2023.1.516","url":null,"abstract":"Portfolio managers attract asset owners by promoting their skills and investment decision process. Performance attribution explains how their decisions add value to portfolios. To be useful, the attribution framework must reflect the decisions made by the managers. Country and currency exposures of global portfolios are often managed with distinct strategies. The author demonstrates that the standard implementation of the Brinson–Fachler attribution framework does not reflect those strategies and produces non-intuitive results. To reflect those strategies, the author shows ways of implementing the Brinson–Fachler framework that incorporate the cost of hedging and the currency surprise and contrasts the results with the standard implementation. He highlights the similarities and the divergences between the Karnosky–Singer and Ankrim–Hensel approaches and proposes a technique to make both approaches equivalent.","PeriodicalId":53670,"journal":{"name":"Journal of Portfolio Management","volume":"49 1","pages":"83 - 99"},"PeriodicalIF":1.4,"publicationDate":"2023-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"46254052","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Factor-based performance attribution is an established practice among quantitative and fundamental strategies alike. Although popular, this attribution technique is not of great help to investors interested in understanding the source of their idiosyncratic performance. Of special interest is the ability to separate stock selection skill and sizing skill. Methods aimed at measuring these skills are inspired by analogies to sports. In this article, the author presents an exact decomposition of the information ratio. The IR is the sum of a breadth-adjusted stock selection skill and a sizing skill. The definition of breadth relies on an intuitive measure of concentration—the Herfindahl Index—rather than on the “square root of n” measure. These quantities can be compared so that portfolio managers can determine the percentage of their realized performance originating from each term. In addition, the decomposition empowers managers to improve their risk-adjusted performance by choosing optimal position sizing, determined by the observed performance of their portfolio. Finally, the decomposition can be further extended to long–short analysis, and provides a natural explanation of an empirical phenomenon: The idiosyncratic performance of the long side of a strategy is often larger than that of the short side.
{"title":"Information Ratio = Selection × Breadth + Sizing","authors":"Giuseppe A. Paleologo","doi":"10.3905/jpm.2023.1.517","DOIUrl":"https://doi.org/10.3905/jpm.2023.1.517","url":null,"abstract":"Factor-based performance attribution is an established practice among quantitative and fundamental strategies alike. Although popular, this attribution technique is not of great help to investors interested in understanding the source of their idiosyncratic performance. Of special interest is the ability to separate stock selection skill and sizing skill. Methods aimed at measuring these skills are inspired by analogies to sports. In this article, the author presents an exact decomposition of the information ratio. The IR is the sum of a breadth-adjusted stock selection skill and a sizing skill. The definition of breadth relies on an intuitive measure of concentration—the Herfindahl Index—rather than on the “square root of n” measure. These quantities can be compared so that portfolio managers can determine the percentage of their realized performance originating from each term. In addition, the decomposition empowers managers to improve their risk-adjusted performance by choosing optimal position sizing, determined by the observed performance of their portfolio. Finally, the decomposition can be further extended to long–short analysis, and provides a natural explanation of an empirical phenomenon: The idiosyncratic performance of the long side of a strategy is often larger than that of the short side.","PeriodicalId":53670,"journal":{"name":"Journal of Portfolio Management","volume":"49 1","pages":"71 - 82"},"PeriodicalIF":1.4,"publicationDate":"2023-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42390758","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The Bayesian statistical method is one of the two major forms of statistical analysis. This approach allows decision makers who have a prior assessment of the probability of some random event occurring to systematically update that probability as more information becomes available. The approach is also used in financial modeling to update the parameters of a model as new data become available. This article reviews the Bayesian methods, discusses the implications for asset management, and describes the limitations of this approach to statistical analysis.
{"title":"Bayesian Methods in Asset Management","authors":"Bradford Cornell","doi":"10.3905/jpm.2023.1.515","DOIUrl":"https://doi.org/10.3905/jpm.2023.1.515","url":null,"abstract":"The Bayesian statistical method is one of the two major forms of statistical analysis. This approach allows decision makers who have a prior assessment of the probability of some random event occurring to systematically update that probability as more information becomes available. The approach is also used in financial modeling to update the parameters of a model as new data become available. This article reviews the Bayesian methods, discusses the implications for asset management, and describes the limitations of this approach to statistical analysis.","PeriodicalId":53670,"journal":{"name":"Journal of Portfolio Management","volume":"49 1","pages":"217 - 223"},"PeriodicalIF":1.4,"publicationDate":"2023-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"47603540","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Natural language processing (NLP) has revolutionized the financial industry, providing advanced techniques for the processing, analyzing, and understanding of unstructured financial text. The authors provide a comprehensive overview of the historical development of NLP, starting from early rules-based approaches to recent advances in deep learning–based NLP models. They also discuss applications of NLP in finance along with its challenges, including data scarcity and adversarial examples, and speculate about the future of NLP in the financial industry. To illustrate the capability of current NLP models, a state-of-the-art chatbot is employed as a co-author of this article.
{"title":"From ELIZA to ChatGPT: The Evolution of Natural Language Processing and Financial Applications","authors":"A. Lo, Manish Singh","doi":"10.3905/jpm.2023.1.512","DOIUrl":"https://doi.org/10.3905/jpm.2023.1.512","url":null,"abstract":"Natural language processing (NLP) has revolutionized the financial industry, providing advanced techniques for the processing, analyzing, and understanding of unstructured financial text. The authors provide a comprehensive overview of the historical development of NLP, starting from early rules-based approaches to recent advances in deep learning–based NLP models. They also discuss applications of NLP in finance along with its challenges, including data scarcity and adversarial examples, and speculate about the future of NLP in the financial industry. To illustrate the capability of current NLP models, a state-of-the-art chatbot is employed as a co-author of this article.","PeriodicalId":53670,"journal":{"name":"Journal of Portfolio Management","volume":"49 1","pages":"201 - 235"},"PeriodicalIF":1.4,"publicationDate":"2023-06-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"47532917","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Equity portfolio tracking error to a benchmark is a most ubiquitous restriction for active portfolios as prescribed by fiduciaries. The restriction is typically a tight range with minimum and maximum ex-ante extremes. The typical capitalization-weighted benchmark (i.e., the S&P 500 Index) has significant changes in diversification character overtime. This brings into question the sensibility of holding tracking error (TE) constant for a skilled active manager. The authors demonstrate that as the concentration of names in the S&P 500 increases, its diversification fades. When this happens, constant tracking error is the enemy of the skilled diversified manager, other things equal. Using multivariate classification and regression trees (CART), they show that high tracking dominates low tracking when index concentration is low, trending lower, and return dispersion is high. Low tracking dominates when the opposite index conditions exist. The authors conclude that the power of a flexible TE process in wealth creation is dominant over inflexibility in the benchmark.
{"title":"The Use and Misuse of Tracking Error","authors":"Eric Sorensen, Nicholas Alonso, D. Bélanger","doi":"10.3905/jpm.2023.1.514","DOIUrl":"https://doi.org/10.3905/jpm.2023.1.514","url":null,"abstract":"Equity portfolio tracking error to a benchmark is a most ubiquitous restriction for active portfolios as prescribed by fiduciaries. The restriction is typically a tight range with minimum and maximum ex-ante extremes. The typical capitalization-weighted benchmark (i.e., the S&P 500 Index) has significant changes in diversification character overtime. This brings into question the sensibility of holding tracking error (TE) constant for a skilled active manager. The authors demonstrate that as the concentration of names in the S&P 500 increases, its diversification fades. When this happens, constant tracking error is the enemy of the skilled diversified manager, other things equal. Using multivariate classification and regression trees (CART), they show that high tracking dominates low tracking when index concentration is low, trending lower, and return dispersion is high. Low tracking dominates when the opposite index conditions exist. The authors conclude that the power of a flexible TE process in wealth creation is dominant over inflexibility in the benchmark.","PeriodicalId":53670,"journal":{"name":"Journal of Portfolio Management","volume":"49 1","pages":"12 - 23"},"PeriodicalIF":1.4,"publicationDate":"2023-06-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42019311","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Tracking a portfolio’s emissions profile over time is a key requirement for any type of climate-aware investment strategy. The challenge in tracking those profiles is that climate metrics are influenced by not only the emissions of companies in the portfolio but also portfolio managers’ decisions, as well as other financial variables such as weights in the portfolio or companies’ enterprise values. In this article, the authors develop an attribution framework that allows investors to disentangle these effects. They focus on financed emissions, which aggregate greenhouse gas emissions “owned” by a portfolio’s holdings, and financed-emissions intensity, which adjusts financed emissions by dividing it by portfolio value. Their approach is to first calculate contributions by looking at changes in a specific input variable while keeping all other input variables constant. Next, they account for effects of simultaneous changes. The results are organized in an attribution tree that allows for a systematic drill-down into the different effects.
{"title":"A Framework for Attributing Changes in Portfolio Carbon Footprint","authors":"Z. Nagy, G. Giese, Xinxin Wang","doi":"10.3905/jpm.2023.1.511","DOIUrl":"https://doi.org/10.3905/jpm.2023.1.511","url":null,"abstract":"Tracking a portfolio’s emissions profile over time is a key requirement for any type of climate-aware investment strategy. The challenge in tracking those profiles is that climate metrics are influenced by not only the emissions of companies in the portfolio but also portfolio managers’ decisions, as well as other financial variables such as weights in the portfolio or companies’ enterprise values. In this article, the authors develop an attribution framework that allows investors to disentangle these effects. They focus on financed emissions, which aggregate greenhouse gas emissions “owned” by a portfolio’s holdings, and financed-emissions intensity, which adjusts financed emissions by dividing it by portfolio value. Their approach is to first calculate contributions by looking at changes in a specific input variable while keeping all other input variables constant. Next, they account for effects of simultaneous changes. The results are organized in an attribution tree that allows for a systematic drill-down into the different effects.","PeriodicalId":53670,"journal":{"name":"Journal of Portfolio Management","volume":"49 1","pages":"163 - 184"},"PeriodicalIF":1.4,"publicationDate":"2023-06-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45536808","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The author presents a framework for specifying and comparing different versions of the calculation of the ex-post “contribution to a return” financial performance measure for each of three levels of application. Employing this framework, the author analyzes a conceptually complete set of versions within each level in order to determine which version best coherently captures the intuitive intent in applying the measure at that level. Specifically, the three levels that the author will explicate are the ex-post contribution of 1) the return of a component of a portfolio on a day to the portfolio’s total return for that day, 2) the total return of a portfolio on a day to a portfolio’s total return for a multi-day period, and 3) the return of a component of a portfolio on a day to a portfolio’s total return for a multi-day period.
{"title":"Various Ex-Post Financial Contributions to a Return and the Different Questions They Address","authors":"Andre Mirabelli","doi":"10.3905/jpm.2023.1.510","DOIUrl":"https://doi.org/10.3905/jpm.2023.1.510","url":null,"abstract":"The author presents a framework for specifying and comparing different versions of the calculation of the ex-post “contribution to a return” financial performance measure for each of three levels of application. Employing this framework, the author analyzes a conceptually complete set of versions within each level in order to determine which version best coherently captures the intuitive intent in applying the measure at that level. Specifically, the three levels that the author will explicate are the ex-post contribution of 1) the return of a component of a portfolio on a day to the portfolio’s total return for that day, 2) the total return of a portfolio on a day to a portfolio’s total return for a multi-day period, and 3) the return of a component of a portfolio on a day to a portfolio’s total return for a multi-day period.","PeriodicalId":53670,"journal":{"name":"Journal of Portfolio Management","volume":"49 1","pages":"101 - 129"},"PeriodicalIF":1.4,"publicationDate":"2023-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49603703","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The use of “information ratios” for benchmark relative (active) returns seems like a small step from the use of the Sharpe ratio for absolute returns. However, something very important got overlooked in that extension. While it is possible (even likely) that all risky assets will outperform the risk-free rate over a sufficiently long horizon, it is impossible for all active managers to outperform sensible benchmarks, even though all active managers (and their investors) must believe they will outperform to rationally pursue active management. Obviously, a material portion of active investors must underperform benchmarks, even though none expects to do so. This failure to accept arithmetic reality is known as the “Central Paradox of Active Management.” This inherent “wrongness” is not reflected in the way an information ratio (IR) is calculated as a simple coefficient of variation, leaving conventional IR values upward biased as performance measures. In this article, the framing of the algebra shows that the degree of bias increases with IR in a nonlinear fashion, so the conventional view that portfolio managers should seek to maximize their information ratio is demonstrably counterproductive.
{"title":"The Central Paradox of Active Management: Maximizing the Information Ratio Is Counterproductive","authors":"Dan Dibartolomeo","doi":"10.3905/jpm.2023.1.509","DOIUrl":"https://doi.org/10.3905/jpm.2023.1.509","url":null,"abstract":"The use of “information ratios” for benchmark relative (active) returns seems like a small step from the use of the Sharpe ratio for absolute returns. However, something very important got overlooked in that extension. While it is possible (even likely) that all risky assets will outperform the risk-free rate over a sufficiently long horizon, it is impossible for all active managers to outperform sensible benchmarks, even though all active managers (and their investors) must believe they will outperform to rationally pursue active management. Obviously, a material portion of active investors must underperform benchmarks, even though none expects to do so. This failure to accept arithmetic reality is known as the “Central Paradox of Active Management.” This inherent “wrongness” is not reflected in the way an information ratio (IR) is calculated as a simple coefficient of variation, leaving conventional IR values upward biased as performance measures. In this article, the framing of the algebra shows that the degree of bias increases with IR in a nonlinear fashion, so the conventional view that portfolio managers should seek to maximize their information ratio is demonstrably counterproductive.","PeriodicalId":53670,"journal":{"name":"Journal of Portfolio Management","volume":"49 1","pages":"25 - 33"},"PeriodicalIF":1.4,"publicationDate":"2023-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44267119","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This article introduces the time-zero direct alpha approach for estimating the outperformance of a private market investment portfolio relative to benchmark(s). To the authors’ knowledge, this is the first published method for private markets to remove the impact of investment timing and accommodate multiple underlying investments with distinct benchmarks in a rigorous manner, without relying on unreliable heuristics. As demonstrated in the article, these problems of investment timing and unobservable subportfolio weights over time can add meaningful noise to estimates of relative performance. This method builds upon the commonly used direct alpha measure for comparing private market returns to public benchmarks. The authors believe that time-zero direct alpha can give private market analysts valuable information for manager selection, portfolio construction, and liquidity planning.
{"title":"Time-Zero Direct Alpha: Investment-Level Calculations for Improved Skill Evaluation","authors":"Nick Keywork, Avi I Turetsky, Barry Griffiths","doi":"10.3905/jpm.2023.1.508","DOIUrl":"https://doi.org/10.3905/jpm.2023.1.508","url":null,"abstract":"This article introduces the time-zero direct alpha approach for estimating the outperformance of a private market investment portfolio relative to benchmark(s). To the authors’ knowledge, this is the first published method for private markets to remove the impact of investment timing and accommodate multiple underlying investments with distinct benchmarks in a rigorous manner, without relying on unreliable heuristics. As demonstrated in the article, these problems of investment timing and unobservable subportfolio weights over time can add meaningful noise to estimates of relative performance. This method builds upon the commonly used direct alpha measure for comparing private market returns to public benchmarks. The authors believe that time-zero direct alpha can give private market analysts valuable information for manager selection, portfolio construction, and liquidity planning.","PeriodicalId":53670,"journal":{"name":"Journal of Portfolio Management","volume":"49 1","pages":"130 - 145"},"PeriodicalIF":1.4,"publicationDate":"2023-06-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"48502304","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
V. Bhansali, Linda Chang, Jeremie Holdom, Matthew Johnson
This article discusses how the traditional approach of measuring performance using time-weighted compounded returns can lead to grossly misleading conclusions in the context of two examples of practical interest. First, the authors show how measuring tail-risk hedging performance using only compounded returns, rather than both returns and timing of cash flows in the context of the underlying portfolio, can lead to erroneous conclusions about the value added by such hedges. Second, they show how measuring performance using compounded returns alone and ignoring timing and size of investment flows can result in contradictory conclusions about the long-term profitability of such investments, using the ARKK exchange-traded fund as an example. They conclude that a more complete approach to performance measurement is essential in order for investors to not be misled by oversimplified metrics, such as compounded returns.
{"title":"Misleading Returns: How Ignoring Cash Flows Can Result in Performance Measurement Errors","authors":"V. Bhansali, Linda Chang, Jeremie Holdom, Matthew Johnson","doi":"10.3905/jpm.2023.1.507","DOIUrl":"https://doi.org/10.3905/jpm.2023.1.507","url":null,"abstract":"This article discusses how the traditional approach of measuring performance using time-weighted compounded returns can lead to grossly misleading conclusions in the context of two examples of practical interest. First, the authors show how measuring tail-risk hedging performance using only compounded returns, rather than both returns and timing of cash flows in the context of the underlying portfolio, can lead to erroneous conclusions about the value added by such hedges. Second, they show how measuring performance using compounded returns alone and ignoring timing and size of investment flows can result in contradictory conclusions about the long-term profitability of such investments, using the ARKK exchange-traded fund as an example. They conclude that a more complete approach to performance measurement is essential in order for investors to not be misled by oversimplified metrics, such as compounded returns.","PeriodicalId":53670,"journal":{"name":"Journal of Portfolio Management","volume":"49 1","pages":"52 - 70"},"PeriodicalIF":1.4,"publicationDate":"2023-06-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"46856348","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"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：