Downside Deviation Research Articles

Perbandingan Value at Risk dan Expected Shortfall pada Portofolio Optimal menggunakan Metode Downside Deviation

Portfolio formation is one of the strategies that investors can do to get the best results Portfolio formation can use the Downside Deviation method. The optimal portfolio with this method uses downside deviation and sets the return below the benchmark as a measure of risk. Every optimal portfolio certainly cannot be separated from risk. To measure risk, you can use the Value at Risk (VaR) and Expected Shortfall values. This study aims to form an optimal portfolio using the Downside Deviation method and continued by comparing the possible losses that occur from the formed portfolio using the VaR and Expected Shortfall values. The data used in this study is the daily closing price data of LQ-45 Index stocks in the banking sector in the period February-June 2023. From the stock data, data selection is carried out by selecting stocks that have a positive expected return and are normally distributed. Then, the optimal portfolio formation stage is continued using the Downside Deviation method and comparing the possible risks formed with the VaR and Expected Shortfall values. The results of this study show that the optimal portfolio with the Downside Deviation method consists of four stocks, namely with the stock codes BRIS.JK, BBRI.JK, BBNI.JK, and BBCA.JK. This study uses a case example by investing capital of Rp100,000,000 with a one-day time period and three levels of confidence, namely 90%, 95%, and 99%. Based on the comparison of the risk value of the portfolio formed using VaR and Expected Shortfall, it is shown that the possible risk with the Expected Shortfall method is greater than the VaR value. Therefore, Expected Shortfall is better in estimating the maximum risk.

The downside deviation quadratic programming for stock portfolio optimisation: an empirical study of shariah and conventional indices in Indonesia

The downside deviation quadratic programming for stock portfolio optimisation: an empirical study of shariah and conventional indices in Indonesia

Enhancing Sustainable Finance through Green Hydrogen Equity Investments: A Multifaceted Risk-Return Analysis

Amidst the global push for decarbonization, green hydrogen has gained recognition as a versatile and clean energy carrier, prompting the financial sector to introduce specialized investment instruments like Green Hydrogen Exchange-Traded Funds (ETFs). Despite the nascent nature of research on green hydrogen portfolio performance, this study examines two key green hydrogen ETFs (i.e., HJEN and HDRO) from April 2021–May 2023, aiming at conducting a multifaceted exploration of their performance, isolating and measuring their sensitivity to the primary market factor, and assessing the capabilities of systematic trading strategies to preserve capital and minimize losses during market downturns. The results spotlight lower returns and higher risks in green hydrogen investments compared to conventional equity (proxied by ETFs offering exposure to developed markets—EFA and emerging markets—EEM) and green energy portfolios (proxied by the ETF ICLN). To comprehensively evaluate performance, an array of risk-adjusted metrics, including Std Sharpe, ES Sharpe, VaR Sharpe, Information ratio, Sortino ratio, Treynor ratio, and various downside risk metrics (historical VaR, modified VaR, Expected Shortfall, loss deviation, downside deviation, and maximum drawdown) are employed, offering a nuanced understanding of the investment landscape. Moreover, single-factor models highlight significant systematic market risk, reflected in notably high beta coefficients, negative alphas, and active premia, underscoring the sensitivity of green hydrogen investments to market fluctuations. Despite these challenges, a silver lining emerges as the study demonstrates the efficacy of implementing straightforward Dual Moving Average Crossover (DMAC) trading strategies. These strategies significantly enhance the risk-return profile of green hydrogen portfolios, offering investors a pathway to align financial and social objectives within their equity portfolios. This research is motivated by the need to provide market players, policymakers, and stakeholders with valuable insights into the benefits and risks associated with green hydrogen investment, considering its potential to reshape the global energy landscape.

Worst-case Omega ratio under distribution uncertainty with its application in robust portfolio selection

AbstractOmega ratio, a risk-return performance measure, is defined as the ratio of the expected upside deviation of return to the expected downside deviation of return from a predetermined threshold described by an investor. Motivated by finding a solution protected against sampling errors, in this paper, we focus on the worst-case Omega ratio under distributional uncertainty and its application to robust portfolio selection. The main idea is to deal with optimization problems with all uncertain parameters within an uncertainty set. The uncertainty set of the distribution of returns given characteristic information, including the first two orders of moments and the Wasserstein distance, can handle data problems with uncertainty while making the calculation feasible.

A New Portfolio Optimization Model Under Tracking-Error Constraint with Linear Uncertainty Distributions.

Enhanced index tracking problem is the issue of selecting a tracking portfolio to outperform the benchmark return with a minimum tracking error. In this paper, we address the enhanced index tracking problem based on uncertainty theory where stock returns are treated as uncertain variables instead of random variables. First, we propose a nonlinear uncertain optimization model, i.e., uncertain mean-absolute downside deviation enhanced index tracking model. Then, we give the analytical solution of the proposed optimization model when stock returns take linear uncertainty distributions. Based on the solution, we find that tracking portfolio frontier is a continuous curve composed of at most n-1 different line segments. Furthermore, we give the condition that tracking portfolio return and risk increase with benchmark return and risk, respectively. Finally, we offer some experiments and show that our proposed model is effective in controlling the tracking error.

Downside deviation quadratic programming for stock portfolio optimisation: an empirical study of Shariah and conventional indices in Indonesia

Downside deviation quadratic programming for stock portfolio optimisation: an empirical study of Shariah and conventional indices in Indonesia

Investment factor timing: Harvesting the low-risk anomaly using artificial neural networks

We perform investment factor timing based on risk forecasts exploiting the low-risk anomaly. Among various risk measures, we find downside deviation most suited for this task. We apply Long Short Term Memory Artificial Neural Networks (LSTM ANNs) to model the relationship between macro-economic as well as financial market data and the downside deviation of factors. The LSTM ANNs allow for complex, non-linear long-term dependencies. We use LSTM-based forecasts to select high- and low-risk factors in setting up an investment strategy. The strategy succeeds in differentiating positive from negative yielding factor investments, and an accordingly constructed investment strategy outperforms every factor individually as well as LASSO and Multilayer Perceptron neural network benchmark models.

Word Portfolio Optimization in the Environment of Zero Interest Rate

This paper provides a deep analysis of ten globally diversified portfolios, composed of different financial instruments: bonds, shares, ETF’s, commodities, indexes, currencies, constructed applying various optimization techniques. Statistical moments, such as mean, standard deviation, kurtosis and skewness of portfolios are compared and discussed. Moreover, performance of the portfolios within the time horizon of one year estimating Sharpe ratio, Treynor ratio, Sortino ratio is presented. Furthermore, a risk analysis of created portfolios is evaluated in terms of historical VaR and CVaR applying confidence interval 95%. The main results of this paper reveal that the portfolio, which is optimized to minimize VaR produces high expected shortfall. Secondly, the Risk Parity portfolio, despite reducing volatility, has delivered the highest kurtosis of the return, which may indicate the possible tail loss. Furthermore, the maximum Sharpe ratio portfolio has delivered extremely high kurtosis in comparison with the kurtosis of the other portfolios. Finally, it is observed that for the Naïve diversification portfolio it has been typical to have the highest downside deviation. 

Application of Data Science for Understanding Emotional Dimensional Behavior and Their Connection to Uncertainty and Risk Behavior

This paper explores multidisciplinary application of emotions and human psychology within market behavior to analyse how emotional dimensions, extracted from Twitter messages and related to cryptocurrency market, are connected to future uncertainty and risk exposure. Although Twitter messages are often used to derive sentiment scores that are then linked to market performance, specific values of emotional components have not been utilised in the previous academic literature to analyze risk behavior. We connect VAD (valence, arousal and dominance) dimensions with the future hourly absolute value of returns, future 24-hour return standard deviation, future 24-hour downside deviation and future 24-hour maximal drawdown (all measuring market uncertainty and risk) on the cryptocurrency market. Results show that all target variables have various predictability from VAD dimensions, where lagged dominance variable (perceived level of control) is the key driver of predictability. Additionally, VAD dimensions have been grouped into distinct clusters by using the k-Means approach. A comparison of selected clusters on in-sample and out-of-sample data showed consistent predictability of identified clusters to all target variables. Results show that emotional components of human emotions, derived from cumulative Twitter messages, actually predict future uncertainty and risk with consistent clustering profile from lagged dominance VAD dimension where lower dominance values predict higher future risk and vice versa.

The Effectiveness of Hedge Fund Investment Strategies under Various Market Conditions

The main purpose of the article was to analyze the effectiveness of the basic investment strategies used by hedge funds in the long term (years 1994-2015) and during the global financial crisis (years 2007-2009). Using information from commercial databases we attempted to verify the hypothesis that alternative hedge funds, regardless of the type of strategy they use, are capable of achieving better outcomes than other capital allocation options, at any time and under a variety of market conditions. We analyzed hedge funds effectiveness in two stages. For stage one, we performed a profitability analysis for the whole hedge funds sector in the two periods and compared the results with the stock rates of return achieved by investors in the global market for the same period and with the risk-free rate. At stage two we calculated ratios that included rate of return and risk, though only for specific strategies. We used traditional portfolio performance measures (Sharpe, Treynor, Jensen ratios) as well as the newer ones (the Sortino ratio, downside deviation). The results show no confirmation that investments carried out by hedge funds are more risky than traditional capital investment methods. Risk associated with investments by the analyzed entities was lower not only in times of prosperity, but also during crisis, providing a clear indication of the higher effectiveness of entities operating in the alternative investment sector and regardless of the changes taking place in the financial market, the length of the capital investment period or the investment strategy being pursued.

Investment Factor Timing: Harvesting The Low-Risk Anomaly Using Artificial Neural Networks

We perform investment factor timing based on risk forecasts exploiting the low-risk anomaly. Among various risk measures, we find downside deviation most suited for this task. We apply Long Short Term Memory Artificial Neural Networks (LSTM ANNs) to model the relationship between macro-economic as well as financial market data and the downside deviation of factors. The LSTM ANNs allow for complex, non-linear long-term dependencies. We use LSTM-based forecasts to select high- and low-risk factors in setting up an investment strategy. The strategy succeeds in differentiating positive from negative yielding factor investments, and an accordingly constructed investment strategy outperforms every factor individually as well as LASSO and Multilayer Perceptron neural network benchmark models.

Teasing Out Tail Risk or, Quantum Mechanics to the Rescue: A Practical Method for Characterizing Downside Risk in Non-Gaussian Return Distributions

Using equations that arise in quantum mechanics, this paper describes a way to more accurately and efficiently represent non-Gaussian return distributions than the standard method of invoking skewness and kurtosis. Then, it provides a new single intuitive number, defined here as the “crash volatility”, to characterize the true left-tail risk as an alternative to the usual downside deviation. The crash volatility can be fed into a typical mean-variance optimizer, allowing the optimizer to finally “see” the risk effect of the non-Gaussian distribution. An example using Amaranth’s returns before it lost -71% in September, 2006 illustrates how these new techniques caught a much higher level of risk lurking in the data.

CAMELS-Based Supervision and Risk Management: What Works and What Does Not

This paper investigates the effectiveness of CAMELS (Capital Adequacy, Assets Quality, Management Efficiency, Earning Efficiency, Liquidity and Sensitivity to Market Risk) based supervision in risk management of A class commercial banks. The riskiness is measured by Downside Deviation (i. e., volatility of returns below minimum average return) and Standard Deviation of ROA and ROE. Using the Generalized Method of Moments (GMM) in secondary balanced panel data during major financial development (i. e., 2004 to 2018; BASEL-I-II-III) of all 28 commercial banks of Nepal; causal relationship between supervision and risk management has been investigated. The result shows that the commercial banks in Nepal can reduce their downside deviation as well as standard deviation of ROA and ROE by reducing the Non-Performing Loan (NPL), maintaining appropriate liquidity and by increasing management efficiency. Further, results justifies the relevance of risk based supervision adopted by central bank and interest spread set. However, increased capital base has not helped in reducing riskiness of banks. Overall, the study finds that among the six parameters of supervision (i. e., CAMELS), five parameters (i. e., AMELS in the priority order of AMLSE) are capable enough to reduce the riskiness of commercial banks if maintained strictly as guided by the central bank.

Stabilized-cycle strategy for a multi-item, capacitated, hierarchical production planning problem in rolling schedules

Little research has been done on hierarchical production planning systems (HPPS) in the context of rolling schedules with service-level constraints. Here, we adapt the stabilized-cycle strategy, which has initially been created for the master planning level (Meistering and Stadtler in Prod Oper Manag 26:2247–2265, 2017), to a two-level, multi-item, capacitated (short-medium-term) HPPS with demand uncertainty. For each planning level, we present extensions for mixed-integer programming models from literature (CLSP-L, PLSP) and introduce anticipation functions, as well as linking constraints. In a computational study, we analyze the performance of the HPPS with different rolling schedule strategies: the period-based, the order-based, and the stabilized-cycle strategy. It turns out that the stabilized-cycle strategy dominates the period-based strategy for all studied instances. For some instances, the stabilized-cycle strategy even dominates the order-based strategy; while in remaining instances, the stabilized-cycle strategy provides non-dominated solutions with a significant smaller downside deviation from service-level agreements and only a minor increase of costs.

Deviation measure in second‐order stochastic dominance with an application to enhanced indexing

AbstractDeviation measures form a separate class of functionals applied to the difference of a random variable to its mean value. In this paper, we aim to introduce a deviation measure in the second‐order stochastic dominance (SSD) criterion to select an optimal portfolio having a higher utility of deviation from its mean value than that of the benchmark portfolio. A new strategy, called deviation SSD (DSSD), is proposed in portfolio optimization. Performance of the proposed model in an application to enhanced indexing is evaluated and compared to other two standard portfolio optimization models that employ SSD criterion on lower partial moments of order one (LSSD) and tail risk measure (TSSD). We use historical data of 16 global indices to assess the performance of the proposed model. In addition to this, we solved the models on simulated data, capturing the joint dependence between indices using historical data of two data sets, each data set comprising six indices. The simulated data are generated by first fitting the autoregressive moving average–Glosten–Jagannathan–Runkle–generalized autoregressive conditional heteroscedastic model on historical data to determine marginal distributions, and thereafter capture the dependence structure using the best fitted regular vine copula. The portfolios from DSSD model achieve higher excess mean return than TSSD model and lower variance, downside deviation, and conditional value‐at‐risk than LSSD model. Also, portfolios from DSSD model demonstrate higher values of information ratio, stable tail‐adjusted return ratio and Sortino ratio than the other two models. Lastly, DSSD model is observed to produce well‐diversified portfolios compared to LSSD model.

Alternativne mere i dekompozicija performansi portfolija investicionih fondova

In addition to the well-established and most commonly used portfolio performance measures, both in theory and practice - the Sharpe ratio, the Treynor ratio and the Jensen's or alpha index, the financial literature also includes other alternative portfolio measures, such as: two modified versions of the Sharpe ratio - the information ratio and the M2 portfolio performance measure, one modified version of the Treynor ratio - the T2 portfolio performance measure, models that measure the market timing abilities of fund managers - the Treynor-Mazuy and the Henriksson-Merton model and a ratio based on the downside risk and the downside deviation as its measure - the Sortino ratio. The paper aims to inform the investors in the Republic of Serbia about the basic features of the aforementioned portfolio performance measures, as well as to point to the importance of understanding the decomposition of the actual portfolio performance of mutual funds.

Index tracking and enhanced indexing using mixed conditional value-at-risk

Index tracking (IT) and enhanced indexing (EI) are two forms of investment strategies which revolve around the movements of the benchmark index. While IT aims to match the performance of the benchmark index, EI intents to outperform the same. In this paper, we seek to design portfolios for IT and EI problems using mixed conditional value at risk (MCVaR). We propose to use the two tail MCVaR (TMCVaR) measure to track the index. Optimizing TMCVaR is a linear program which minimizes the upper deviation and the downside deviation from the benchmark index and hence meets the objective of IT. On the other hand, we propose a two step procedure for EI problem. In step one, we design a discrete Markov chain model to filter a few stocks on the basis of their high probability of gain over the benchmark index. In step two, we assign optimal weights to the filtered assets through maximizing any of the two variants of the STARR ratio with MCVaR or the STARR ratio with deviation MCVaR (DMCVaR). Maximizing the STARR ratio either with MCVaR or DMCVaR is a linear program and hence tractable. We analyze the empirical performance of the proposed models over 17 world-wide indices using the rolling window approach. We consider two IT and four EI models from the literature for a comparative analysis. It is exhibited that the proposed IT model outperforms the other two IT models over several performance measures including higher correlation value with the benchmark index and lower tracking error, and the two proposed EI models outperform the other four EI models in achieving higher excess mean returns from the benchmark index.

IDENTIFYING VULNERABILITY TO POVERTY: A CRITICAL SURVEY

AbstractIn the economic literature on poverty, various methods have been proposed for measuring a phenomenon known as ‘vulnerability’. However, after more than a quarter century of research, no consensus has been reached on how to identify such vulnerable individuals within a given population. Some misunderstandings have also arisen from the overlapping of other closely related concepts, such as the expectation of being poor, expected poverty, multi‐period poverty and risk exposure. This paper offers a detailed conceptual discussion on vulnerability to poverty and its related elements, reviewing a wide range of identifying criteria provided in the literature. It is found that according to the state of the art in this field of research, two key elements stand out in identifying vulnerable individuals: an expected well‐being below the poverty line and a relevant risk of falling into poverty due to downside deviation from a reference level of well‐being. The traditional classification of vulnerability approaches has been updated into four groups: (i) those that stress the element of exposure to risk; (ii) those that emphasize the element of expected poverty; (iii) those that define vulnerability through a utility gap and (iv) those that are supported by a mean‐risk dominance criterion.

Dynamic Indexes: Equity Rotation and Factor Timing

It is well understood that factor premiums, identified as contributing to portfolios risk and return, are time-varying and conditional on market phases. Besides, Amenc et al. [2016] report alternative weighting schemes not to yield similar performances, based on a common value and small-cap exposure, but identify alternative performance drivers. Thereon, we build on the existing literature regarding the factor exposure of value-weighted and equal-weighted portfolios and make us of the diversity-weighted indexing approach, in order to dynamically derive the degree of tilt towards an equal-weighting. We show that such strategies correspond to a factor rotation, whilst demonstrating favorable after-cost properties regarding risk-adjusted returns, downside deviation, portfolio concentration and tracking error. As such, it fulfills the characteristics of what Lo [2016] refers to as 'dynamic indexes.'

Chasing Two Rabbits: Challenges in Benchmarking Liquid Alternatives

Liquid alternatives are required to compare their performance to a broad-based market index. Despite their objective to offer low correlations with equity returns, these funds provide traditional equity indexes as benchmarks two and a half times more than any other benchmark category. By pursuing strategies unrelated to their primary benchmarks, liquid alternatives can generate significant tracking errors, which cause biased information ratios. We review the challenges of benchmarking liquid alternatives and recommend use of the downside deviation or maximum drawdown. This issue is important, as research shows how performance relative to even a mismatched benchmark is a significant determinant of fund flows.