Portfolio Construction Research Articles

Robust optimization with order statistic uncertainty set

In this paper, we propose a new uncertainty set for robust models of linear optimization problems. We first study data-free and distribution-free statistical properties of continuous and independent random variables using the Probability Integral Transform. Based on these properties, we construct a new uncertainty set by placing constraints on the order statistics of random variables. We utilize the quantiles of random variables to depict the uncertainties and then adopt the formulation of the assignment problem to develop a tractable formulation for the order statistic uncertainty set. We show that the robust optimization models with the interval uncertainty set, the budget uncertainty set, and the demand uncertainty set can be obtained as special cases of the robust optimization model with the order statistic uncertainty set. Finally, using a robust portfolio construction problem as an example, we show via numerical experiments that the order statistic uncertainty set has better performance than other uncertainty sets when the sample size is small and the correlation between random variables is low.

An on-line machine learning return prediction

This paper introduces a novel methodology for predicting relative asset returns using a large dataset. Our approach utilizes on-line universal portfolio construction and generates a closed-form prediction formula based solely on historical data. Our results demonstrate that the predictive error can be as low as 2% and is robust. These findings suggest that on-line machine learning techniques have the potential to predict relative asset returns when sufficient data is available.

Alpha Strategy for Stripping Beta Returns for Pure Bond Funds

Contemporarily alpha strategy is favorable by investors and funders for its stable performances among volatility market. Although alpha strategy shows great extra returns, it might have great drawdown on account of the stochastic fluctuations of the market (e.g., hedge funds’ poor performance during COVID-19 early state including Bridge Water). Despite from stocks and futures, the bond funds are rather a low volatility type underlying assets, hence the alpha strategy based on pure bond will be more favorable for risk aversion investors. By building a five-factor model, this paper dissolves the returns of pure bond funds to the five factors of level, slope, convex, credit and default, which are separated from each other, and Alpha, which is not directly explained by the factors, and verifies the effectiveness of this alpha factor for selecting pure bond funds. Taking 2019 to 2022 as the time interval, the group backtest results show that the alpha factor stripped of beta has a good effect on the selection of pure bond funds in China's public offering fund market, and the base selection ability is better than the Sharpe Ratio and other traditional risk-return factors. Overall, these results shed light on guiding further exploration of portfolio construction with high alpha based on bond funds.

Factor Investing Webinar

In this webinar, Frank Fabozzi moderated a discussion with four prominent quantitative investment professionals. The experts were asked about their approaches to factor investing, the characteristics they look for in a factor, and which factors they believed in and why. The webinar covered topics such as the viability of value as a factor over the past decade, the role of ESG as a factor, and the possibility of discovering new factors. The difference between factors and signals was discussed, along with the potential for factor timing. The panelists also examined how to incorporate factors into portfolios, identified the biggest challenges facing factor investing and shared the main findings from research on factor portfolio construction. They also discussed the most promising areas of research in the field, including Machine Learning and the application of factors to private markets. Additionally, the discrepancy between the use of factors in equities versus fixed income was discussed. Overall, the webinar provided a comprehensive overview of the challenges and opportunities associated with factor investing.

Comparative Analysis of Industry and Country Diversification Benefits of the Equity Portfolio

This study focuses on the construction of investment portfolios using different optimization methods (1/N, risk parity, hierarchical risk parity, mean variance) based on two ways of asset diversification—cross-country and cross-industry—and assesses their effectiveness using a number of metrics. The main research question is which approach to asset diversification turns out to be the most profitable for the investor in terms of return and risk. Research results’ degree of susceptibility of fluctuations in time is assessed by an interperiod analysis with the division of the time interval under consideration into crisis and noncrisis years. According to the results, the industry asset allocation strategy outperforms the asset allocation strategy focused on country diversification. In crisis years, these differences are amplified, but the results are highly dependent on the portfolio optimization method.

Penerapan Model Markowitz dalam Memperoleh Portofolio Optimal di Pasar Saham IDX30

Investment in financial market is inextricably linked to risk. The risk cannot be eliminated, but indeed it can be minimized. By using Markowitz model, investor can gain optimal result while taking on acceptable risk. This study intends to examine the findings of optimal portfolio construction using Markowitz model on IDX 30. The method was quantitative descriptive. The population was all companies listed on IDX 30. Then, there were three companies chosen as the sample. The investigation of the three stocks yielded two efficient portfolios. First were the stocks of PT Bank Negara Indonesia, Tbk. and PT Astra International. The second was stocks of PT Bank Negara Indonesia, Tbk. and PT Adaro Energy Indonesia, Tbk. Each stock provided a return of 1.03 % with the risk level of 7.75%, and 4.4% with the risk level of 8.49%. The conclusion is that by using these two portfolios, investor can get optimal expected returns with the degree of risk comparable to individual equities. Thus, it is suggested for the investor to apply the calculation of Markowitz model on Indonesia Stock Exchange over a longer period than the one used in this study. It is also suggested for them to select the portfolio in accordance to their preferences.

Portfolio Construction based on AMZN, JPM, CSCO, GM and TSLA

Contemporarily, the sudden and rapid spread of the Newcastle pneumonia outbreak around the world caused stock markets in various countries to experience dramatic volatility in the early stages of the outbreak. The outbreak was an extreme stress test for global capital markets. On this occasion, investors change the risk strategy with high volatility to stable portfolio. On this basis, this paper will select relevant data of five listed companies in the United States in the last seven years through python, covering the period before and after the outbreak of New Canopy pneumonia. The aim of the study is to try to establish an optimal portfolio through CAPM model and asset allocation theory in order to maximize the acquisition return under investment risk diversification, facilitate investors to carry out practical exercises, further analyze the experimental results and discuss the limitations. Overall, these results shed light on guiding further exploration of portfolio construction under pandemic.

Constructing An Optimal Portfolio Based on Markowitz And Nonlinear Programming Methods

Portfolio construction is extremely essential for the trading of stocks construct an optimal portfolio is one of the frequently asked questions in the financial sector as it is directly related to the profits of investors. While some people use financial theory only in their investments and do not combine it with mathematical methods. So, we explore how to construct an optimal portfolio through financial and mathematical models and theories based on the analysis of stock prices of some Chinese companies in the past year. We find that the stock of Xi'an Catering Co., Ltd and Joyvio Group Limited is extremely attractive for us, at 91 percent and 6 percent of our cash invested respectively. The Sharpe ratio is 0.09948, which means the reward of unit risk is 0.09948. So, we need to invest more in stocks with higher expected returns compared to others. Because it could bring more profits for us. Usually, the companies with an increased stock price during a period will satisfy the requirements as they have better development potential. Our study can also be improved by using the Fama-French three factors model and Machine learning.

Portfolio Construction Based on Combination of Individual Stocks and LSTM

With the development of the idea of portfolio optimization, it has become one of the important topics in the field of modern finance for achieving the balance between maximizing the return of assets and minimizing the risk. This study selected ten stocks from WIND and used LSTM Neural Network as the fundamental forecasting model. Based on the method of cyclic prediction, we predicted the asset’s return forecast value of the next day by rolling multi-dimensional historical features. Five stocks with the best outcomes were selected for optimization, using both the return component and the volatility, covariance matrix of the stocks’ returns as the covariance matrix to build the efficient frontier and to optimize the weighting. The Monte Carlo Simulation was used to build the maximum Sharpe ratio portfolio and minimum volatility portfolio. The maximum Sharpe ratio portfolio showed that the Midea and Mindray had the potential to profit dramatically at the risk of high volatility. However, relatively high volatility means the possibility of profiting is questionable, so a Minimum Volatility Portfolio was also conducted, which indicated conservative investors usually reduce the risk of their investment as much as possible. Whereas the max Sharpe ratio portfolio showing that risk-seeking investors are more willing to take risks if it can generate more profit. Overall, a hybrid intelligent algorithm based on cyclic prediction, Monte Carlo Simulation and artificial neural network was proposed, and the new applications of machine learning and mathematical methods in asset allocation were promoted.

Hellinger’s distance to normal distribution as market invariant

Main purpose of distance based portfolio constructions is for portfolio imitation. Here we used distance from normal distribution for other purpose. We attempted to find static market invariant within possible linear combinations of given random variables. We conjectured that “closeness” to normal distribution of possible portfolios in market may reliably represent market microstructure with possible correlations between assets. Taking the squared Hellinger’s distance, we sought for each level of desired mean return the portfolio whose return distribution is closest to Gaussian, with variance taken from efficient frontier found by initially solving mean-variance problem. We found that minimal distance differs significantly from market to market. The sensitivity check showed small average sensitivity for 5% change of a 5% portion of data, small sensitivity for adding new variables in simulated market, and extreme sensitivity to bin numbers. Though distance to normality differed among markets, its sensitivity being small enough in average sometimes showed extreme changes.

An integrated framework for classification and selection of stocks for portfolio construction: Evidence from NSE, India

Investment extortion in the stock market is a crucial aspect considered by the investors. Therefore, investors implemented different strategies. This study was intended at constructing an investment portfolio (IP) of stocks within the NSE 100 listed companies of Non-parametric nature, fulfilling the basic premise of portfolio making that is, reducing risks while yielding an attractive return higher than any other instrument for the investors. Using DP omnibus test, the desired sample of companies following the non-normal distribution was achieved. Using financial beta, we have selected the outcome based on the nature of their ‘return’ and ‘risk'. We introduce TOPSIS (Technique for order of performance by similarity to ideal solution), a multi-criteria decision-making process (MCDM) to study the profitability of stocks, rank wise for each year, and finally, the Bayes portfolio model help to select the overall profitability associate with low risk for the construction of the portfolio.

Performance comparison of multifractal techniques and artificial neural networks in the construction of investment portfolios

This work aims to compare the performance of the traditional portfolios of the S&P500, Markowitz, and Sharpe with the multifractal trend fluctuation portfolios (MF-DFA) and portfolios of artificial neural networks with Student's asymmetric probability classification (ANN-t). In this study, we use daily data for S&P500 stocks between January 18, 2018, and July 12, 2022, where we backtest return and risk metrics such as annual volatility, Value at Risk, Sharpe Ratio, Sortino Ratio, Beta, and Jensen´s Alpha. For both return and risk, we obtain the results confirming that the ANN-t technique might indicate better investment entries, which contradicts the Efficient Market Hypothesis (EMH).

When Safe-Haven Asset Is Less than a Safe-Haven Play

AbstractWe propose a four-state regime-switching model that pairs low-volatility and high-volatility (HV) states to test eight stock–safe-haven asset portfolios’ risk properties. We find the correlations between gold, U.S. T-bond, and the Swiss franc and stock markets are negative or zero in all states, including the HV–HV state, while the correlations between Bitcoin (BTC) and stock markets are positive in the HV–HV state, implying that gold, T-bond, and the Swiss franc are full safe-havens and BTC is a partial safe-haven asset. Moreover, our model is effective in portfolio construction, performing better than conventional time-varying generalized autoregressive conditional heteroskedasticity-based models.

Looking Back at the Past 25 Years and Forward to the Next 25

Earlier in the second half of the 20th century, asset management evolved dramatically as it became common practice to revisit certain principles which had underpinned the institutional world: the introduction of quantitative practices in general and the “discovery” of the efficient frontier in particular forced a rethink of risk-taking and thus portfolio construction. Index investing probably is the single most visible development still visible today. The private wealth world waited longer to embark on the same path of reformation, but made up for its initial lateness by innovating at a fast pace since the early 1990s. Three crucial changes accompanied a gradual shift from the inherited wealth, as the main client, to the advent of new wealth, which was arguably more prepared to ask for new approaches. The recognition that taxes matter and that tax day is everyday was, chronologically, the first, though it capitalized in part on work that had been carried out in the institutional world with certain insurance companies. The advent of open architecture simply reflected a reality which already existed in the institutional world, but was not necessarily appreciated by managers who were “product-driven.” This allowed a broadening of the strategies offered to individuals, including the use of alternative and illiquid strategies. Finally, the recognition of the fact that individuals have multiple goals, multiple time horizons and different risk tolerances for different goals brought Goals-Based Wealth Management, with a key catalyst being the work of Das, Markowitz, Scheid and Statman in 2010. Looking ahead, one can see that a lot still has to be done, with an important element probably involving the integration of insurance into the wealth management mix.

Robust Portfolio Optimization Strategies in The Serbian Stock Market

Research Question: This paper investigates the performances of six portfolios constructed using robust optimization methods in the Serbian stock market. Motivation: Motivated by the lack of research that analyses the allocation strategies based on robust optimization in the other non-US markets, this paper analyses the ability of these strategies to produce positive performance in the Serbian financial market. Idea: This paper aims to check whether robust strategies can provide positive risk-adjusted performance compared to simple strategies. Data: The analysis was performed on daily data from 2017 to 2020. Tools: We used monthly portfolio rebalancing with an estimation period of 24 months, applying budget and no-short selling constraints in portfolio construction. As a benchmark, we used two simple strategies, the strategy of market index replication and the equal weighting strategy (1/N). Consequently, the performance of the portfolios is evaluated once a month and calculated for the entire investment period. Findings: Empirical results suggest that robust optimization methods improve portfolio performance on a risk-adjusted basis. The increase in performance is affected by an increase in turnover, so the stability of weights in the portfolio depends on the compliance of the model characteristics with the conditions prevailing in the market. Contribution: To the best of our knowledge, this is the first article that analyses the performance of robust optimization portfolios for the Serbian stock markets. Analysing the performance of robust optimization strategies and comparing them to two simple strategies, this paper contributes to the existing literature by checking their possibility of obtaining a positive performance in less developed markets. Additionally, all information presented in this paper could help investors optimize their risk allocation and profitability.

Equity-Market-Neutral Strategy Portfolio Construction Using LSTM-Based Stock Prediction and Selection: An Application to S&P500 Consumer Staples Stocks

In recent years, a great deal of attention has been devoted to the use of neural networks in portfolio management, particularly in the prediction of stock prices. Building a more profitable portfolio with less risk has always been a challenging task. In this study, we propose a model to build a portfolio according to an equity-market-neutral (EMN) investment strategy. In this portfolio, the selection of stocks comprises two steps: a prediction of the individual returns of stocks using LSTM neural network, followed by a ranking of these stocks according to their predicted returns. The stocks with the best predicted returns and those with the worst predicted returns constitute, respectively, the long side and the short side of the portfolio to be built. The proposed model has two key benefits. First, data from historical quotes and technical and fundamental indicators are used in the LSTM network to provide good predictions. Second, the EMN strategy allows for the funding of long-position stocks by short-sell-position stocks, thus hedging the market risk. The results show that the built portfolios performed better compared to the benchmarks. Nonetheless, performance slowed down during the COVID-19 pandemic.

Does reinforcement learning outperform deep learning and traditional portfolio optimization models in frontier and developed financial markets?

Advancements in machine learning have opened up a wide range of new possibilities for using advanced computer algorithms, such as reinforcement learning in portfolio risk management. However, very little evidence has been provided on the superior performance of reinforcement learning models over traditional optimization models following the mean-variance framework in different financial market settings. This study uses two experiments with data from the Vietnamese and U.S. securities markets to justify whether advanced machine learning models could outperform traditional portfolios' cumulative returns while optimizing the Sharpe ratio. The results suggest that reinforcement learning consistently outperforms the established methods and benchmarks in both experiments, even when using a very similar degree of diversification in portfolio construction and the same input data. This study confirms the ability of reinforcement learning to provide dynamic responses to market conditions and redefine the risk-return standard in the financial system.

Portfolio Optimization Using Minimum Spanning Tree Model in the Moroccan Stock Exchange Market

Portfolio optimization is a pertinent topic of significant importance in the financial literature. During the portfolio construction, an investor confronts two important steps: portfolio selection and portfolio allocation. This article seeks to investigate portfolio optimization based on the Minimum Spanning Tree (MST) method applied on the Moroccan All Shares Index (MASI) historical stock log returns covering the period from 2 January 2013 to 27 October 2022 allowing us to build two portfolios: MST-Portfolio and MST-Portfolio 2. Portfolio selection was carried out for MST-Portfolio and MST-Portfolio 2, respectively, based on 63 stocks or using the Degree Centrality (DC) measure and portfolio allocation for both portfolios was carried through the use of the Inverse Degree Centrality Portfolio (IDCP). The obtained portfolios were compared with the Minimum Variance Portfolio (MV Portfolio) and Equal Weighting Portfolio (EW Portfolio) using centrality measures, diversification, and backtesting. According to the used indicators analysis, MST-Portfolio and MST-Portfolio 2 are the most well-performed and robust portfolios showing a good performance during the studied period, even during the COVID-19 crisis, and ensuring a good level of diversification. The findings demonstrate that both suggested methods can enhance portfolio performance, evidence that can help investors or active managers when optimizing their portfolios.

Application of Three Commonly Used Portfolios in the Hong Kong Stock Market

This paper will use stocks in Hang Seng Index to figure out the construction of portfolios. In addition, Fama experiment would be used to eliminate the idiosyncratic risk and R language would be regarded as the tool to do onerous computation. And 9 stocks randomly selected plus one risk-free asset would be entailed to construct portfolios, and finally all the stocks without missing values would be entailed. This paper may shed light on investing combination of stocks of Hang Seng index or some others about how to assign the weight proportion to reach different purposes.

Portfolio Construction and Analysis Based on the Hang Seng Index from 2009 to 2021

Maximizing returns at a low risk is a popular goal for investors. The Hang Seng Index (HSI) is a significant indicator of Hong Kong stocks, which is made up of 66 stocks traded on the Hong Kong stock market. This study aims to analyze and the find optimum portfolio based on the Hang Seng index. For this purpose, Fama experiment are applied firstly to evaluate the systematic risk and the idiosyncratic risk. But for convenience, three stocks from three different companies, Tencent, BYD and the CITIC are chosen to construct the portfolio. Then R is used in this study to analyze the portfolio. The minimum variance portfolio, Sharpe ratio and the tangency portfolio are all calculated by R then. The results show that the weights of the tangency portfolio are -200% for Tencent, 42% for BYD and 258% for CITIC, which may be helpful to make the investment decision.