Portfolio Construction Methodology Research Articles

Network Risk Parity: graph theory-based portfolio construction

AbstractThis study presents network risk parity, a graph theory-based portfolio construction methodology that arises from a thoughtful critique of the clustering-based approach used by hierarchical risk parity. Advantages of network risk parity include: the ability to capture one-to-many relationships between securities, overcoming the one-to-one limitation; the capacity to leverage the mathematics of graph theory, which enables us, among other things, to demonstrate that the resulting portfolios is less concentrated than those obtained with mean-variance; and the ability to simplify the model specification by eliminating the dependency on the selection of a distance and linkage function. Performance-wise, due to a better representation of systematic risk within the minimum spanning tree, network risk parity outperforms hierarchical risk parity and other competing methods, especially as the number of portfolio constituents increases.

Diversified Spectral Portfolios: An Unsupervised Learning Approach to Diversification

The question of how to diversify an investment portfolio is one with many possible answers. Over the past couple of years, the industry and academic literature have been shifting focus from an asset-driven answer to a factor-driven one, sparking special interest in the use of implicit factors identified through unsupervised learning. However, issues around the stability and implementation of these, in the context of diversification, have left a gap between what is an academic exercise and what is an implementable methodology. This article aims to fill this gap by presenting a diversification-focused portfolio construction methodology that takes advantage of singular value decomposition to identify implicit factors and uses hierarchical agglomerative clustering to address some of the challenges surrounding its implementation. In out-of-sample Monte Carlo simulations, this methodology can provide better risk-adjusted performance than other commonly used portfolio diversification approaches.

Measuring and Optimizing the Risk and Reward of Green Portfolios

We study the performance of green portfolios in both the US and Chinese markets, constructed using a broad range of climate-related environmental metrics, including carbon emissions, water consumption, waste disposal, land and water pollutants, air pollutants, and natural resource use. We compare several popular long-only and long–short green portfolio construction methodologies and find that a method based on Treynor–Black weights offers the most robust performance, thanks to its ability to quantify alphas for individual assets using only a small number of parameters. In the United States, green portfolios (e.g., low-carbon portfolios) have realized positive alphas in excess of Fama–French factors, a significant portion of which can be explained by an unexpected increase in climate concerns over the past decade, rather than positive expected returns. In contrast, Chinese investors have borne a cost for holding green assets instead of brown assets over the past seven years, implying a positive carbon premium, the opposite of US markets.

Sharpe Parity Redux

In this article, the authors investigate the performance of the Sharpe parity asset allocation strategy relative to the more well-known 60/40, mean–variance, risk parity, and min-variance strategies. Each portfolio selection strategy was tested under a number of different constraints and asset class configurations. In addition to historical data, bootstrapped simulated time series were used to test the robustness of the analysis. The primary conclusion was that the performance of each of the strategies considered was highly dependent on the constraints applied during the portfolio construction process. As such, portfolio managers choosing to implement a given portfolio construction methodology must be careful in choosing their asset universe and calibrating their asset class level constraints to give their portfolios the maximum performance advantage versus rival portfolio selection methodologies and manager benchmarks.

Factor Construction Zoo: Are Factor Exposures Created Equal?

The answer is no. Factor investing provides investors with a low-cost avenue to participate in stock selection. Investors earn excess returns for taking on factor risk, which is often measured by factor exposure. However, the relation between the return and factor exposure is nonlinear. Large-scale simulation shows that similar target factor exposures can be engineered using various portfolio construction methodologies, but the resulting portfolios exhibit significant dispersion in expected returns and co-movement with the market and across factor funds. The dispersion increases with target factor exposures. As such, some factor exposures are more efficient than others. This article further studies a comprehensive list of portfolio construction choices for value, momentum, and quality funds; discusses the trade-off at work; and provides a framework for the assessment of factor exposure efficiency. It is important to account for nonlinearity in constructing or evaluating factor funds.

Revisiting Quality Investing

In the field of factor investing, quality is undoubtedly the equity factor with the weakest consensus. This research investigates the best way to define it. In order to capture the multi-faceted reality of the factor depicted in academia, we address the quality factor through a multidimensional process by defining four self-reliant pillars: profitability, earnings quality, safety and investment. To better fit institutional investor's' needs, we analyze the resulting factor by focusing on the last eighteen years and on a global developed markets universe of liquid stocks (large- and mid-caps).In a long-short framework, our quality factor delivers a statistically significant alpha that cannot be explained by loadings on conventional equity factors (market, value, size and momentum). Most regions and dimensions display positive contribution to this alpha, with the noticeable exceptions of the Eurozone region and the safety dimension. In a long-only framework, our quality factor outperforms its benchmark by 2.8% per annum over the entire analysis period, with an information ratio of 0.81. Furthermore, the outperformance has been very consistent since the 2008 Global Financial Crisis (GFC). The four dimensions are weakly correlated with each other and are therefore complementary. We show that safety is of particular importance during periods of market turmoil (GFC, Covid-19 pandemic) and that the dimension is therefore part of the quality factor in its own right. On the Eurozone side, a sector-neutral portfolio construction seems to be more suited.We also introduce a new portfolio construction methodology by implementing a clustering approach based on the K-means algorithm to group together companies based on features that are related to both fundamentals and market characteristics. This approach allows to capture dynamic variations between fundamentals and other stock features. This fully implementable process results in better quality factor performance without impacting the associated risk measures or the portfolio’s quality exposure, as measured on the unconstrained quality factor.

Risk Capacity Portfolio Construction

This paper introduces a portfolio construction framework for risk-averse investors that aim to meet or exceed a client’s capital accumulation needs for a future event, such as retirement. Risk Capacity Portfolio Construction (RCPC) presents a risk-optimized alternative to Target Date Funds (TDFs). Risk Capacity Portfolio Construction is an economic sciences innovation that is validated by Skew-Risk Modeling, which was first presented in Targeted Return Portfolio Construction (Sherwood, 2019) and is detailed within this paper in its application to retirement-focused investors. Risk Capacity Portfolio Construction factors skewness and kurtosis into the risk management and asset allocation of an individual’s life cycle and can optimize risk beyond simply accounting for the equity risk premium and human capital. TOPICS:Portfolio theory, wealth management, retirement, risk management, portfolio construction, options, equity portfolio management, tail risks Key Findings ▪ This paper highlights an alternative portfolio construction methodology to Target Date Funds (TDFs) in order to better prepare individuals for retirement, on both a risk-adjusted and absolute basis. ▪ This paper highlights an investor’s ability to incorporate skewness and kurtosis within the asset allocation process of a portfolio that is intended to prepare an individual or group for a future dated event. ▪ This paper highlights how the implied risk probabilities within the options market, referred to herein as Tail Risk Ratios (TRR), can be incorporated in the investment process.

Factor Hunting: Are Factor Exposures Created Equal?

The answer is no. Investors hunt factor exposures and premium across the stock universe. However, the relation between factor exposures and returns is non-linear. Large-scale simulation shows that similar target factor exposures can be engineered using various portfolio construction methodologies, but the resulting portfolios exhibit significant dispersion in expected returns and comovement with the market and other factor funds. The dispersion increases with target factor exposures. As such, some factor exposures are more efficient than others. This paper further studies a comprehensive list of portfolio construction choices for value, momentum, and quality funds, discusses the tradeoff at work, and provides a framework for the assessment of factor exposure efficiency. It is important to account for non-linearity in constructing or evaluating factor funds.

Next-Generation Commodity Benchmarks: RASATM Benchmarks Designed to Align with CIO Investment Objectives

Investors can construct commodity benchmarks better aligned with their investment objectives. This is important because common “plain-vanilla” benchmarks, constructed to mimic relative production activity, may be inconsistent with a CIO’s objectives. CIOs can use the Real Asset Sensitivity Analysis (RASA TM) framework to construct a benchmark that meets a desired investment objective (target an inflation or growth beta) while accounting for beta estimation uncertainty. To do so, investors specify a desired inflation or growth exposure, at a given investment horizon, then construct a RASA benchmark using individual commodity futures indices. Such benchmarks not only would have delivered better risk-adjusted performance historically, compared to traditional commodity benchmarks, but would have been better aligned with investors’ investment objectives. Today’s index technology enables a CIO to construct a RASA commodity benchmark, with independent third-party calculation and reporting, to track the commodity exposure that better matches their investment objectives. We illustrate how this can be done using our portfolio construction methodology.

Trend-Following versus Cross-Sectional Momentum: A Data-Driven Statistical Factor Comparison

The persistence of returns for an asset (momentum effect) is well known and robust: Practitioners use a variety of different approaches and methodologies to capture it. They often distinguish between trend-following (or time series) and cross-sectional momentum models. This article presents a comparison of the two methodologies based on statistical factors. For that purpose, we create trend-following and cross-sectional models in four asset classes (Equity, Forex, Interest Rate, and Commodity). We show that the first have relied on a dynamic (time varying) exposure to a few dominant statistical factors to generate performance. Interestingly, the predominant factor can easily be interpreted as the exposure to the asset class. On the other hand, cross-sectional models are more diversified but historically underperformed trend-following models. They, however, yield interesting results in more heterogeneous asset classes (for instance, Commodity and Interest Rate). Finally, we show that the categorization is of minor importance, as returns are driven by exposure to factors, which can be modified in multiple ways, such as, for example, by changes in the portfolio construction methodology. TOPICS: Quantitative methods, portfolio construction, factor-based models Key Findings • Trend-following models by asset class have relied on a dominant statistical factor to generate performance, which can be interpreted as exposure to the asset class. • Cross-sectional models are more diversified and appear to be more interesting in heterogeneous asset classes as they tend to hedge the dominant statistical factor. • The categorization in cross-sectional and trend-following models is of minor importance, as they might both load on similar performance drivers depending on portfolio constraints or portfolio construction methodologies.

Cryptocurrencies and the low volatility anomaly

This study examines the low volatility anomaly in the cryptocurrency market. Constructing long-short portfolios for a sample of 1000 cryptocurrencies for the period April 28, 2013 – November 1, 2019, we find no evidence of a significant low volatility premium. This result is in contrast to the empirical findings from the equity, bond, and commodity markets and contributes to the debate on the efficiency of cryptocurrencies. In contrast to earlier studies, we find that the cryptocurrency market is far more efficient than expected, even after controlling for different sample sizes, rebalancing periods and / or portfolio construction methodologies.

Risk Management and the Optimal Combination of Equity Market Factors

Managing the intertemporal risk of optimally constructed multifactor portfolios adds to performance. The increases in Sharpe ratios are in addition to the utility that investors gain from controlling how much active risk they are exposed to over time. We derive a simple closed-form formula for security weights in optimal multifactor portfolios with an active-risk target. We test the risk control of five well-known factors—value, momentum, small size, low beta, and profitability—and the optimal multifactor portfolio. Our empirical research was carried out on the large-capitalization US equity market for 1966 through 2019. We conclude that for the equity market, more active factors are better than fewer if each subportfolio is “pure” as to factor, anchored to the benchmark, and combined on the basis of forecastable risks. Our portfolio construction methodology allows for transparent performance attribution and replication of the process in other markets and time periods.

The Devil Is in the Details: The Risks Often Ignored in Low-Volatility Investing

With increasing investor interest in low-volatility equity strategies comes a need for greater scrutiny of different methodologies used to achieve low-volatility exposure. In an earlier article, the authors investigated the analytical differences between a variety of approaches to constructing low-volatility portfolios. In this article, the authors turn their attention to the empirical differences between common approaches to low volatility. They find that a traditional optimizer-based approach to building low-volatility portfolios has large sensitivities to the risk inputs used in the process. In fact, using the same portfolio construction methodology but changing the risk inputs even slightly can lead to large differences. The magnitude of this sensitivity should give investors pause; even across risk inputs in which differences are valid, variations persist and can be harmful to portfolio performance. The authors show that there are other, more robust, ways of achieving low-volatility portfolios without this input sensitivity (e.g., risk balancing) and suggest that investors should consider this lack of input sensitivity as a valuable characteristic in low-volatility investing. TOPICS:Volatility measures, exchanges/markets/clearinghouses, risk management Key Findings • Using an optimizer to gain exposure to the low-volatility anomaly/premiums subjects the portfolio construction process to input sensitivity, which can lead to wide variation in results. • Input sensitivity affects not only the portfolio’s sector and stock positioning but also its risk exposures. This variation in risk exposures can lead to very different outcomes for investors. • There are more robust ways to gain exposure to the low-volatility anomaly/premiums, and the authors show that a risk-balanced approach not only captures the anomaly/premiums well but does so under different risk model assumptions.

Dynamic Risk Management of Equity Market Factors

Managing the intertemporal risk of optimally constructed multifactor portfolios adds to performance. The increases in Sharpe ratios are in addition to the utility that investors gain from controlling how much active risk they are exposed to over time. We derive a simple closed-form formula for security weights in optimal multifactor portfolios with an active-risk target. We test the risk control of five well-known factors—value, momentum, small size, low beta, and profitability—and the optimal multifactor portfolio. Our empirical research was carried out on the large-capitalization US equity market for 1966 through 2019. We conclude that for the equity market, more active factors are better than fewer if each factor subportfolio is “pure” as to factor, anchored to the benchmark, and combined on the basis of forecastable risks. Our portfolio construction methodology allows for transparent performance attribution and replication of the process in other markets and time periods.

Climate Change Investment Risk: Optimal Portfolio Construction Ahead of the Transition to a Lower-Carbon Economy

There is an increasing likelihood that governments of major economies will act within the next decade to reduce greenhouse gas emissions, probably by intervening in the fossil fuel markets through taxation or cap \& trade mechanisms (collectively ``carbon pricing''). We develop a model to capture the potential impact of carbon pricing on fossil fuel stocks, and use it to inform Bayesian portfolio construction methodologies, which are then used to create what we call Smart Carbon Portfolios. We find that investors could reduce ex-post risk by lowering the weightings of some fossil fuel stocks with corresponding higher weightings in lower-risk fossil fuel stocks and/or in the stocks of companies active in energy efficiency markets. The financial costs of such de-risking strategy are found to be statistically negligible in risk-return space. Robustness of the results is explored with alternative approaches.

Climate change investment risk: optimal portfolio construction ahead of the transition to a lower-carbon economy

There is an increasing likelihood that governments of major economies will act within the next decade to reduce greenhouse gas emissions, probably by intervening in the fossil fuel markets through taxation or cap & trade mechanisms (collectively “carbon pricing”). We develop a model to capture the potential impact of carbon pricing on fossil fuel stocks, and use it to inform Bayesian portfolio construction methodologies, which are then used to create what we call Smart Carbon Portfolios. We find that investors could reduce ex-post risk by lowering the weightings of some fossil fuel stocks with corresponding higher weightings in lower-risk fossil fuel stocks and/or in the stocks of companies active in energy efficiency markets. The financial costs of such de-risking strategy are found to be statistically negligible in risk-return space. Robustness of the results is explored with alternative approaches.

Active Risk Budgeting: A Portfolio Construction Methodology for Futures Strategies

We present a portfolio construction methodology for futures strategies that incorporates active trading and also borrows salient features from the risk-parity methodology. We document the evolution of expected risk and return based portfolio construction methodologies and propose a new methodology which corrects the crucial assumption in risk parity that all investments have similar risk adjusted returns and improves on Grinold and Kahn’s position sizing methodology by using a optimization based framework. We demonstrate a baseline implementation of active risk budgeting that improves upon both mean variance optimization and risk budgeting and document the challenges associated with its implementation for futures trading.

Active Risk Budgeting is better than the Tangency Portfolio

In this paper, we show empirically that Active Risk Budgeting is a superior portfolio construction methodology to the tangency portfolio method postulated by Mean Variance Optimization. We compare the performance of Active Risk Budgeting and Tangency Portfolio in a series of systematic experiments as we gradually increase the predictive accuracy of the input signal. We find that almost always Active Risk Budgeting has better returns. Only when the signal used for portfolio construction perfectly knows the Sharpe Ratio of the securities in the next five days does the performance of the tangency portfolio catch up to Active Risk Budgeting. Given the results, we would recommend the use of Active Risk Budgeting in portfolio construction for active investment strategies on derivatives.

O Alinhamento Estratégico para a confecção do portfólio de projetos no caso do Planejamento Estratégico 2017-2018 da Assembleia Legislativa do Rio Grande do Norte

O presente artigo tem como objetivo descrever o processo de escolha, de portfólio dos projetos estratégicos, realizada no Planejamento Estratégico 2017-2018 da Assembleia Legislativa do Estado do Rio Grande do Norte (ALRN), Brasil. Metodologicamente, a pesquisa qualitativa caracterizou-se por um estudo de caso de múltiplas fontes de evidência (Yin, 2016), com três fases: entendimento de sistemáticas, compilação dos dados quanto à formação de portfólio e análise do produto final, a fim de esclarecer como ocorreu o processo de escolha do portfólio de projetos no Planejamento Estratégico 2017-2018, bem como traçar um perfil de formação de portfólio em relação ao alinhamento estratégico almejado pela ALRN. Os resultados apontaram para um alinhamento estratégico onde foram identificadas as fases de definição de eixos estratégicos, critérios de descarte e vinculação dos catorze projetos na criação do portfólio de projetos estratégicos. Contudo, registra-se, como ponto de atenção, a inclusão de projetos diretamente pela direção da instituição na fase final de definição do portfólio, fato este de risco para a manutenção do alinhamento, por não seguir a metodologia geral de construção do portfólio. Assim, o artigo visa contribuir nas discussões sobre o tema, ao partir de uma perspectiva de alinhamento estratégico onde se acrescenta uma discussão acerca da seleção dos projetos, de modo a garantir a devida prioridade dentro da organização, buscando a eficiência e eficácia em seus processos.

Practical Applications of Not All Factor Exposures Are Created Equal

Practical Applications Summary In Not All Factor Exposures Are Created Equal, from the 2018 Quantitative Special Issue of The Journal of Portfolio Management, authors Eric Sorensen, Mark Barnes, Nick Alonso, and Edward Qian (all of PanAgora Asset Management in Boston, MA) evaluate the use of different portfolio construction methodologies for managing factor portfolios. They consider four types of portfolios that hold the factor exposure constant and have varying security weighting schemes: factor weighted (FW), cap weighted (CW), equal weighted (EW), and risk parity weighted (RP). The find that using a risk-aware weighting scheme exemplified by their RP portfolios generally achieves both strong upside participation and an advantageous defensive posture when a factor underperforms. They ascribe the advantage of the RP approach to sector positions, particularly in emphasizing exposure to defensive sectors such as consumer staples, healthcare, and utilities.