Portfolio Allocation Strategies Research Articles

Whether ARIMA-based Portfolio Can Beat the Market Index

Portfolio optimization is a long-term topic in the financial field, which can maximize returns while minimizing risks. It is widely used in production and daily life, such as stock investment, production optimization, and engineering models. This article selects data from 10 stocks, namely KO, PG, PEP, CL, MDLZ, STZ, PM, KMB, GIS, and K, from January 24, 2024, to June 14, 2024. Firstly, this article uses the ARIMA model to learn the first 70% of the data and predict the last 30% of the data, and then uses portfolio strategy to evaluate the results. This article obtained the maximum Sharpe ratio model on the Monte Carlo effective boundary. By combining the maximum Sharpe ratio model with the SP 500 index within the same time frame, this article concludes that the market performs better. This article provides different portfolio allocation strategies for risk averse investors seeking stable positive returns in turbulent markets.

Interconnections and contagion among cryptocurrencies, DeFi, NFT and traditional financial assets: Some new evidence from tail risk driven network

For digital assets (cryptocurrencies, DeFi tokens and NFTs) and traditional financial assets such as stock indices and commodities, this paper established a dynamic downside risk network to explore the network aggregating structure, important nodes and the tail risk spillover between asset groups. This study reveals that there exist tail risk spillover between digital financial assets and traditional financial assets, where the risk spillover between the digital assets and the commodities is bidirectional and the risk spillovers from the stocks to the digital assets as well as the commodities to stocks are unidirectional. For important assets, the NFT Decentraland and Brazil’s BVSP stock index play central roles in the network, while the FTSE stock index and gold exhibit strong hedging potential against global downside risk spillovers. Furthermore, we consider the impact of uncertainty in US economic policy and US stock volatility on tail risk contagion, which is the network aggregation level. Regarded as exogenous shocks, the US stock circuit breaker in March 2020 and the US Federal Reserve rate rising in March 2022 were found to significantly increase the tail risk contagion, especially the risk spillover from the stocks market to the digital financial market. Moreover, by using the TVP-SV-VAR model, we find that increases in the US EPU index and the US VIX index can both exacerbate the tail-risk contagion between digital and traditional financial assets. This expands the influencing factors of risk contagion between financial assets. The empirical findings carry important implications not only for portfolio managers and investors in portfolio allocation and investment strategies between digital and traditional financial assets, but also for government regulators in risk management and economic policy formulation.

Beyond hype: Unveiling the herd effect in ESG and non-ESG cryptocurrency portfolios

This paper investigates herding behavior in the cryptocurrency market, specifically comparing environmental, social, and governance (ESG)-designated assets and non-ESG assets. It addresses a research gap concerning investor behavior towards ESG and non-ESG cryptocurrencies. Our study reveals that investor herding behavior manifests in specific years and in response to significant events, such as El Salvador’s adoption of Bitcoin. Notably, the findings suggest the presence of intentional herding behavior in 2023 data for ESG and non-ESG cryptocurrencies. These insights offer valuable guidance for investors and policymakers seeking to mitigate risks and optimize portfolio allocation strategies.

Quantile coherency across bonds, commodities, currencies, and equities

This paper examines quantile coherency in bonds, commodities, currencies, and equities using a novel quantile coherency approach. While recent literature has explored single-frequency tail- and time-frequency dependence in asset returns, we provide fresh evidence on asset return dependence across quantiles (proxying business cycles or market conditions) at different frequencies (representing investment horizons). Considering sixty-seven individual asset return series in four asset classes, we observe that low frequency (yearly) dependence is stronger in the bond, foreign exchange, and equity markets. Specifically, we find strong dependence between the German and French bond markets, heating oil and crude oil, gold and silver, British Pound, and Euro, French and German and Canadian and US equities. As we report asset return interdependence in different business cycles and at different time horizons, these results have important implications for portfolio allocation and investment strategy formulation.

Portfolio Optimization by LSTM with a Selection of Six Stocks

As the financial industry undergoes continuous evolution, efficient asset allocation has become increasingly crucial. However, traditional methods employed for portfolio optimization are often deemed inefficient and in need of improvement. To address this, recent advancements in deep learning techniques provide a promising perspective to tackle portfolio optimization, offering new possibilities for maximizing returns or minimizing risk based on specific objectives and constraints. This study delves into the analysis of stock data from six distinct industries. By utilizing the LSTM model and employing the Monte Carlo method, efficient frontier, and other advanced techniques, a training set is constructed to generate predictions using the first 80% of the data. For testing purposes, the remaining 20% of the data is utilized to assess how well the created portfolio performed. Various performance metrics such as portfolio returns, volatility, Sharpe Ratio, and maximum reduction are calculated to assess the effectiveness of the LSTM-based portfolio. Additionally, a comparison is made against other benchmark portfolios or strategies. The results for evaluation show that the LSTM-based portfolio outperform the commonly used benchmark model. This study illuminates the potential of deep learning in the financial industry, presenting groundbreaking applications that offer novel portfolio allocation strategies.

Portfolio Optimization Strategy Based on Four Deep Learning Models

Deep learning techniques have provided a fresh outlook on the evergreen subject of portfolio optimization within the finance domain. This article selects the stocks of Google, Tesla, Tractor Supply Company, Analog Devices, and Duke Energy Corporation and deploys four deep learning models to estimate returns and covariance respectively. The mean-variance model is utilized to generate the target portfolio for each deep learning model, incorporating the predicted outcomes. Ultimately, the returns of each portfolio are compared to the market benchmark (S&P 500) returns. The findings demonstrate that the proposed target model outperforms the market benchmark (S&P 500) across multiple financial metrics. This study highlights the groundbreaking and promising applications of deep learning in the financial sector, providing valuable insights into innovative portfolio allocation strategies for risk-averse investors who aim to achieve stable and positive returns even in turbulent market conditions.

A Profitable Portfolio Allocation Strategy Based on Money Net-Flow Adjusted Deep Reinforcement Learning

A Profitable Portfolio Allocation Strategy Based on Money Net-Flow Adjusted Deep Reinforcement Learning

Spillover effects from China and the United States to Key Regional Emerging Markets: A dynamic analysis

While a limited number of studies have employed wavelet techniques to distinguish between market interdependence and contagion within the context of market co-movement, many of them have primarily relied on conclusions derived from multivariate wavelet analysis. In contrast, this paper introduces a complementary approach based on the univariate wavelet technique, specifically the Maximal Overlap Discrete Wavelet Transform (MODWT). This approach entails assessing the correlation between key market series that have been decomposed into both the time and frequency domains. The proposed technique is applied to evaluate the co-movement and dynamic spillover effects from two prominent advanced markets, namely China and the United States, to six major emerging economies across different regions. The findings obtained through various wavelet techniques consistently support the notion that interdependence is the predominant characteristic in terms of shock spillovers and stock market co-movement between the selected advanced and key emerging markets. Furthermore, the results indicate that both the US and China cycles are synchronized with those of the key emerging economies. These insights are of significant importance for investors and asset managers, offering valuable guidance for asset diversification and portfolio allocation strategies during both tranquil and turbulent periods.

LSTM-based Portfolio Optimization Strategy for SP500

Portfolio optimization is a perennial topic in the field of finance and recent breakthrough in deep learning techniques offers a new perspective to tackle it. This study selects 30 stocks of SP500 in different sectors through various constraints and deploys Long Short-Term Memory and Ledoit-Wolf Shrinkage to estimate returns and covariance respectively. The target portfolio is then obtained by inputting the predicted results into the mean-variance model, which is dynamically updated on a daily basis given evolving market information. The results show that the target model this study proposed surpasses the market benchmark (SP500), 1/N portfolio, and other mean-variance variants in terms of numerous financial metrics. Moreover, the target model exhibits volatility invariance and the capability to mitigate risk while extracting returns. This study showcases the revolutionary and promising applications of deep learning in the financial industry, shedding light on novel portfolio allocation strategies for risk-averse investors seeking stable positive returns in turbulent markets.

Application of ARIMA Model in Portfolio Optimization

This paper investigates the portfolio allocation strategy using the ARIMA time series model with eight companies in four industries: Technology, Healthcare, Financial Services, and New Energy. This research aims to identify the best portfolio allocation approach that maximizes the returns for investors in each industry. This paper collected daily stock prices of Apple and Microsoft in the technology industry, Johnson & Johnson and Pfizer in the healthcare industry, JPMorgan Chase and Goldman Sachs in the financial services industry, and Tesla and NextEra Energy in the new energy industry from January 2016 to December 2019. The logarithmic returns first were calculated for the first three years as the training data. Then, an appropriate ARIMA time series model would be fitted to forecast the stock return for the last year and perform the portfolio optimization based on the predicted return for 2019 following a result of weights for each asset. By comparison, the paper also applied portfolio optimization to the raw logarithmic daily return from 2016 to 2018 without the time series model fitted and got another set of weights for each asset. Finally, both cumulative returns for these two portfolios computed and compared with visualizations. There is a finding that the modified strategy of portfolio with ARIMA model forecasted data could provide a more obvious trade-off relationship between volatility and sharpe-ratio for returns than the traditional one performed with the raw data. The result of this paper provides a potential scope of an application of portfolio optimization worked with time series modeling for investors to make investment decisions that possibly yield a higher return.

Portfolio allocation strategy for active learning Kriging-based structural reliability analysis

Recently, numerous studies have focused on structural reliability analysis, with the Kriging-based active learning method being particularly popular. A variety of Kriging-based learning functions have been proposed, and shown to perform well in various tasks. However, no single learning function has been demonstrated to consistently outperformed the others in all tasks, and selecting the most appropriate learning function for a given task remains a challenge in engineering applications. In this paper, inspired by the multi-armed bandit strategy, a portfolio allocation of different learning functions is proposed to resolve the issue of selecting a single one, where the better learning functions are selected online according to their past performance. Finally, three classical numerical examples and two engineering applications are adopted to validate the effectiveness of the proposed method.

A hybrid level-based learning swarm algorithm with mutation operator for solving large-scale cardinality-constrained portfolio optimization problems

We propose a hybrid variant of the level-based learning swarm optimizer (LLSO) for solving large-scale portfolio optimization problems. This solver fills the gap due to the inadequacy of the particle swarm optimization algorithm for high-dimensional instances. We aim to extend the classical mean-variance formulation by maximizing a modified version of the Sharpe ratio subject to cardinality, box, and budget constraints. The algorithm involves a projection operator to deal with these three constraints simultaneously. Further, we implicitly control transaction costs thanks to a rebalancing constraint handled by a suitable exact penalty function. In addition, we develop an ad hoc mutation operator to modify candidate exemplars in the highest level of the swarm. The experimental results, using three large-scale data sets, show that including this procedure improves the accuracy of the solutions. Then, a comparison with other variants of the LLSO algorithm and two state-of-the-art swarm optimizers points out the outstanding performance of the proposed solver in terms of exploration capabilities and solution quality. Finally, we assess the profitability of the portfolio allocation strategy in the last five years using an investable pool of 1119 constituents from the MSCI World Index.

High-Dimensional Distributionally Robust Mean-Variance Efficient Portfolio Selection

This paper introduces a novel distributionally robust mean-variance portfolio estimator based on the projection robust Wasserstein (PRW) distance. This approach addresses the issue of increasing conservatism of portfolio allocation strategies due to high-dimensional data. Our simulation results show the robustness of the PRW-based estimator in the presence of noisy data and its ability to achieve a higher Sharpe ratio than regular Wasserstein distances when dealing with a large number of assets. Our empirical study also demonstrates that the proposed portfolio estimator outperforms classic “plug-in” methods using various covariance estimators in terms of risk when evaluated out of sample.

Re-evaluating portfolio diversification and design using cryptocurrencies: Are decentralized cryptocurrencies enough?

This paper investigates the portfolio diversification potential of a pool of cryptocurrencies classified based on their degree of leadership. We employ the mean-variance and the higher-order moments optimization approaches to evaluate the diversification potential of centralized and decentralized cryptocurrencies across multiple frameworks. While theoretical implications of the mean-variance and the higher-order moments optimization approaches are similar, our results suggest that the latter provides a more precise portfolio allocation strategy because it considers investor risk-aversion for each moment. Furthermore, we find that extending the pool of cryptocurrencies achieves marginal diversification benefits due to considerable co-movements among the cryptocurrencies. Moreover, we find that decentralized cryptocurrencies offer greater diversification potential than centralized cryptocurrencies, although centralized cryptocurrencies carry some diversification potential during alt-seasons. In order of their weights, Bitcoin, Chainlink, and Ethereum (all decentralized) offer the highest contribution to portfolio diversification across most portfolio frameworks, while Ethereum offers greater diversification benefits during the alt-seasons.

Bank Portfolio Allocation Strategy and Its Probability of Failure: Case of the Russian Banking Sector Purge

This paper aims to discover portfolio allocation strategies that facilitate a bank’s stability. The paper examines the phenomenon of massive failures of Russian banks in the period from 2013 to 2019, in order to identify which of the banks’ strategic decisions regarding assets and liabilities, as well as portfolio structure, lead to higher stability. The dataset contains financial indicators and prudential ratios of 895 commercial banks operating in Russia during that period. 507 banks, or 57% of all banks, lost their license during the considered period. Cases of bank failures were classified depending on whether the Central Bank identified any illegal activities conducted by the failed bank. The high failure rate provides an opportunity to study the differences between failed and non-failed banks in order to determine the factors associated with lowerfailure probability. Following the approach applied in most of the previous studies, we use a logistic regression to model the effect of different asset and liability portfolio structure on the failure probability. The hypothesis that failure probability of a bank is affected by its strategic focus of forming an assets and liabilities portfolio was statistically confirmed. We found that the focus of a bank’s activity on providing loans to individuals and attracting deposits from companies leads to lower failure probability, confirming the results of previous studies. Also, we found that more active cooperation with other banks in terms of both borrowing and lending is associated with lower failure probability. Furthermore, we found that banks are less likely to borrow from or lend money to their fellow banks that later fail with illegal activity accusations.Finally, we found that unlike the EU banks, Russian banks with higher profitability ratios are more stable. The results are relevant for industry practitioners in facilitating the development of a more resilient bank strategy, as well as for regulators for incorporation in early warning models.

Optimising portfolio diversification and dimensionality

A new framework for portfolio diversification is introduced which goes beyond the classical mean-variance approach and portfolio allocation strategies such as risk parity. It is based on a novel concept called portfolio dimensionality that connects diversification to the non-Gaussianity of portfolio returns and can typically be defined in terms of the ratio of risk measures which are homogenous functions of equal degree. The latter arises naturally due to our requirement that diversification measures should be leverage invariant. We introduce this new framework and argue the benefits relative to existing measures of diversification in the literature, before addressing the question of optimizing diversification or, equivalently, dimensionality. Maximising portfolio dimensionality leads to highly non-trivial optimization problems with objective functions which are typically non-convex and potentially have multiple local optima. Two complementary global optimization algorithms are thus presented. For problems of moderate size and more akin to asset allocation problems, a deterministic Branch and Bound algorithm is developed, whereas for problems of larger size a stochastic global optimization algorithm based on Gradient Langevin Dynamics is given. We demonstrate analytically and through numerical experiments that the framework reflects the desired properties often discussed in the literature.

Financial turbulence, systemic risk and the predictability of stock market volatility

This paper adds a novel perspective to the literature by exploring the predictive performance of two relatively unexplored indicators of financial conditions, i.e. financial turbulence and systemic risk, over stock market volatility using a sample of seven emerging and advanced economies. The two financial indicators that we utilize in our predictive setting provide a unique perspective on market conditions, as they relate directly to portfolio performance metrics from both volatility and co-movement perspectives and, unlike other macro-financial indicators of uncertainty, or risk, can be integrated into diversification models within forecasting and portfolio design settings. Since the data for the two predictors are available at a weekly frequency, and our focus is to produce forecasts at the daily frequency, we use the generalized autoregressive conditional heteroskedasticity-mixed data sampling (GARCH-MIDAS) approach. The results suggest that incorporating the two financial indicators (singly and jointly) indeed improves the out-of-sample predictive performance of stock market volatility models over both the short and long horizons. We observe that the financial turbulence indicator that captures asset price deviations from historical patterns does a better job when it comes to the out-of-sample prediction of future returns compared with the measure of systemic risk, captured by the absorption ratio. The outperformance of the financial turbulence indicator implies that unusual deviations in not only asset returns, but also in correlation patterns play a role in the persistence of return volatility. Overall, the findings provide an interesting opening for portfolio design purposes, in that financial indicators, which are directly associated with portfolio diversification performance metrics, can also be utilized for forecasting purposes, with significant implications for dynamic portfolio allocation strategies.

Liquidity risk and stock returns: empirical evidence from industrial products and services sector in Bursa Malaysia

This study investigates the impact of liquidity risk on stock returns of 149 firms in the industrial products and services sectors of Bursa Malaysia from January 2000 to December 2018 with a monthly frequency dataset. This study employed the two-stage standard procedures in asset pricing to estimate the significant effect of liquidity risk on industrial products and services stock returns. The results show that the investors require liquidity premium for stocks whose illiquidity co-moves with market illiquidity and market return while shifting their investment to liquid stocks when the market becomes illiquid, thus positive premium for stocks whose return is higher during the illiquid market. It suggests that two liquidity risks, namely commonality in liquidity and the covariances between stock illiquidity and market returns, and aggregate liquidity risk explain the cross-sectional returns variations across stocks in the industrial products and services sector, thus partly support the LCAPM model. We provide evidence on the important role of liquidity risks on the cross-sectional industrial products and services stock returns in Bursa Malaysia in the LCAPM framework. The findings of this study may be useful for investment decision-making and portfolio allocation strategy under the liquid and illiquid securities conditions. For policymakers, understanding the impact of liquidity risks on stock returns for the industrial products and services sectors may help enhance market liquidity for economic growth. Therefore, our findings contribute to the practical and policy implications.

Interconnected banks and systemically important exposures

We study the impact of the interplay between the structure of the financial network and market conditions on financial stability in the European banking system. We capture two channels of financial contagion. The first channel concerns direct interconnectedness, via a network of interbank loans, banks’ loans to other corporate and retail clients, and securities holdings. The second channel concerns indirect interconnectedness, via overlapping exposures to common asset classes. To assess the impact of contagion, we apply the structural valuation model NEVA, in which common shocks to banks’ external assets are reflected in a consistent way in the market value of banks’ mutual liabilities through the network of obligations. Applying the model to a unique supervisory data set collected by the European Central Bank that covers 26 large banks in the euro area we identify a strongly non-linear relationship between diversification of exposures, shock size, and losses due to interbank contagion. We also demonstrate the potential for contagion effects to amplify first-round stress test results due to interconnectedness. Finally, we provide insights into the potential impact of more diversified versus more domestic portfolio allocation strategies on the propagation of contagion, which are relevant to policy discussions about international risk sharing, for instance, in the context of the EU Capital Market Union.

Chaos based portfolio selection: A nonlinear dynamics approach

Time series forecasting is of fundamental importance for financial market prediction and, consequently, for portfolio allocation strategies. However, non-stationarity and non-linearity of most financial time series often make these tasks difficult to perform. In this paper, we propose a methodology based on chaos and dynamical systems theory for non-linear time series forecasting and investment strategy development, which is able to correctly make predictions at long time horizons. We construct Constant Chaoticity Portfolios (CCP) and evaluate their performances on the survival components of the STOXX Europe 50 index and the Hang-Seng index. Results show that the CCP overwhelms several competing alternatives, both in terms of net profits and risk-return profiles. Our findings are confirmed by a sensitivity analysis on the parameters of the underlying model and over different choices of forecast horizons.