Optimal Portfolio Research Articles

Package CovRegpy: Regularized covariance regression and forecasting in Python

Abstract This paper will outline the functionality available in the CovRegpy package which was written for actuarial practitioners, wealth managers, fund managers, and portfolio analysts in the language of Python 3.11. The objective is to develop a new class of covariance regression factor models for covariance forecasting, along with a library of portfolio allocation tools that integrate with this new covariance forecasting framework. The novelty is in two stages: the type of covariance regression model and factor extractions used to construct the covariates used in the covariance regression, along with a powerful portfolio allocation framework for dynamic multi-period asset investment management. The major contributions of package CovRegpy can be found on the GitHub repository for this library in the scripts: CovRegpy.py, CovRegpy_DCC.py, CovRegpy_RPP.py, CovRegpy_SSA.py, CovRegpy_SSD.py, and CovRegpy_X11.py. These six scripts contain implementations of software features including multivariate covariance time series models based on the regularized covariance regression (RCR) framework, dynamic conditional correlation (DCC) framework, risk premia parity (RPP) weighting functions, singular spectrum analysis (SSA), singular spectrum decomposition (SSD), and X11 decomposition framework, respectively. These techniques can be used sequentially or independently with other techniques to extract implicit factors to use them as covariates in the RCR framework to forecast covariance and correlation structures and finally apply portfolio weighting strategies based on the portfolio risk measures based on forecasted covariance assumptions. Explicit financial factors can be used in the covariance regression framework, implicit factors can be used in the traditional explicit market factor setting, and RPP techniques with long/short equity weighting strategies can be used in traditional covariance assumption frameworks. We examine, herein, two real-world case studies for actuarial practitioners. The first of these is a modification (demonstrating the regularization of covariance regression) of the original example from Hoff & Niu ((2012). Statistica Sinica, 22(2), 729–753) which modeled the covariance and correlative relationship that exists between forced expiratory volume (FEV) and age and FEV and height. We examine this within the context of making probabilistic predictions about mortality rates in patients with chronic obstructive pulmonary disease. The second case study is a more complete example using this package wherein we present a funded and unfunded UK pension example. The decomposition algorithm isolates high-, mid-, and low-frequency structures from FTSE 100 constituents over 20 years. These are used to forecast the forthcoming quarter’s covariance structure to weight the portfolio based on the RPP strategy. These fully funded pensions are compared against the performance of a fully unfunded pension using the FTSE 100 index performance as a proxy.

Research on Portfolio Optimization Model based on Machine Learning Algorithm in Stock Market

Research on Portfolio Optimization Model based on Machine Learning Algorithm in Stock Market

The Maximal and Minimal Distributions of Wealth Processes in Black–Scholes Markets

The Black–Scholes formula is an important formula for pricing a contingent claim in complete financial markets. This formula can be obtained under the assumption that the investor’s strategy is carried out according to a self-financing criterion; hence, there arise a set of self-financing portfolios corresponding to different contingent claims. The natural questions are: If an investor invests according to self-financing portfolios in the financial market, what are the maximal and minimal distributions of the investor’s wealth on some specific interval at the terminal time? Furthermore, if such distributions exist, how can the corresponding optimal portfolios be constructed? The present study applies the theory of backward stochastic differential equations in order to obtain an affirmative answer to the above questions. That is, the explicit formulations for the maximal and minimal distributions of wealth when adopting self-financing strategies would be derived, and the corresponding optimal (self-financing) portfolios would be constructed. Furthermore, this would verify the benefits of diversified portfolios in financial markets: that is, do not put all your eggs in the same basket.

Optimization of investment portfolios of Chinese commodity futures market based on complex networks

Abstract China commodity futures market network is constructed. Commodity is the node of the network, and the network link is defined by the price correlation matrix. We analyze the relationship between the centrality of each commodity in the commodity futures market network and the optimal weight of the commodity portfolio, empirically examine the market system factors and commodity personalized factors that affect the centrality of commodity, and evaluate the effect of network structure on the optimization of commodity portfolio selection under the mean-variance framework. It is found that the commodities with high network centrality are often related to industrial products and have high volatility. Commodities with higher centrality have lower portfolio weights. We put forward a kind of commodity futures investment strategy based on network, according to the network centricity grouping the commodities, the network centricity lower edge of the commodity structure of the portfolio, cumulative yield is better than that of centricity higher core product portfolio, the whole market portfolio yield, but due to large maximum retracement, lead to the stability and ability to resist risk compared with the other two groups of goods combination. The main contribution of this paper is to optimize portfolio selection by establishing the relationship between portfolio weight and commodity centrality by using commodity futures market network as a tool.

Optimal selection of manufacturing enterprise financialization based on simulated annealing investment portfolio theory

The global market competition is becoming increasingly fierce, and manufacturing enterprises need to invest and expand. However, the traditional financial optimization of manufacturing enterprises has faced problems such as low efficiency and inaccurate search for the optimal solution, which has made manufacturing enterprises likely to face financial risks. The investment portfolio can enable enterprises to obtain the maximum profit on a certain risk level, or reduce their investment risk on a certain return level as far as possible. If the combinatorial optimization is realized, it can be applied to the optimal selection of manufacturing enterprises’ financialization. This article analyzed the respective characteristics of Genetic Algorithm (GA) and Simulated Annealing (SA) algorithms, and analyzed the combination of GA and SA algorithms to solve the optimal investment portfolio through GA-SA algorithm, thereby helping manufacturing enterprises to make the optimal choice for financialization. The experimental results of this article indicated that the GA-SA algorithm solved the problem of GA algorithm easily falling into local optima, SA algorithm’s initial temperature and generation mechanism, and improved the efficiency of finding the optimal solution. Meanwhile, the experimental results showed that the average optimal solutions of Genetic Algorithm, Simulated Annealing algorithm, and GA-SA algorithms for 36 stock portfolios in Enterprise 1 were 69, 69, and 107, respectively. The average optimal solutions of the three algorithms for 36 stock portfolios in Enterprise 2 were 73, 90, and 112, respectively. This proves that the number of optimal solutions searched by GA-SA algorithm is higher than that of GA and SA algorithm, and also proves that it is effective to use GA-SA algorithm to optimize investment portfolio and help manufacturing enterprises to make optimal financial choices.

Prospect theory-based formulation of chance constrained portfolio optimization problem using loan

Prospect theory-based formulation of chance constrained portfolio optimization problem using loan

MCN portfolio: An efficient portfolio prediction and selection model using multiserial cascaded network with hybrid meta-heuristic optimization algorithm

ABSTRACT Generally, financial investments are necessary for portfolio management. However, the prediction of a portfolio becomes complicated in several processing techniques which may cause certain issues while predicting the portfolio. Moreover, the error analysis needs to be validated with efficient performance measures. To solve the problems of portfolio optimization, a new portfolio prediction framework is developed. Initially, a dataset is collected from the standard database which is accumulated with various companies’ portfolios. For forecasting the benefits of companies, a Multi-serial Cascaded Network (MCNet) is employed which constitutes of Autoencoder, 1D Convolutional Neural Network (1DCNN), and Recurrent Neural Network (RNN) is utilized. The prediction output for the different companies is stored using the developed MCNet model for further use. After predicting the benefits, the best company with the highest profit is selected by Integration of Artificial Rabbit and Hummingbird Algorithm (IARHA). The major contribution of our work is to increase the accuracy of prediction and to choose the optimal portfolio. The implementation is conducted in Python platform. The result analysis shows that the developed model achieves 0.89% and 0.56% regarding RMSE and MAE measures. Throughout the analysis, the experimentation of the developed model shows enriched performance.

Analyzing Nexus between Crude Oil, Gold, Dollar and Equity Markets with Structural Break: ARDL Evidence from India

This paper explores the price relations between the Crude oil, Gold, US dollar, and equities during several economic episodes. The long- and short-term causal relationships between asset markets are examined in this study. The research uses a ‘Vector Error Correction Model (VECM)’ and an ‘Autoregressive Distributed Lag (ARDL)’ bounds test to examine monthly data from April 1999 to March 2022. The results demonstrate the weak connections between asset markets throughout the different sample periods. Furthermore, the integration of most markets is uneven and changes over time. The currency and equity markets are adjusting to the long-run equilibrium only at a slow pace. We suggest that systematic risk factors must be taken into account while jointly modeling market linkages. This study improves on previous research in the subject by demonstrating the time-varying effects of asset price links on portfolio optimizations in different economic episodes.

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.

Dynamic consumption and portfolio choice considering information learning and stochastic interest rate

This paper investigates the consumption–portfolio choice of investors with information learning ability when faced with historical performance, within a stochastic interest rate framework. We simultaneously analyze investor behaviors, specifically momentum trading and contrarian trading. Using the dynamic programming method, we derive the closed-form solutions. The numerical results demonstrate how investors with heterogeneous characteristics make consumption–investment allocations in response to different historical performances. We find that momentum investors tend to overreact to short-term market fluctuations, while contrarian investors focus on long-term investment potential and market reversals. In addition, our findings suggest that information utility is time-sensitive and decays over time.

Analysing Malaysian Technology Sector during COVID-19: A Minimum Spanning Tree Approach

This paper examines the structural changes of the companies listed in the technology sector in Bursa Malaysia during the COVID-19 pandemic. The closing price of each stock is used as the basis for calculating the cross-correlation between the companies' returns. Minimum spanning trees are employed to analyse the relationships and visualize the correlation changes for each period. Additionally, degree centrality is utilized to analyse the network's structure further. This study covers the period from 2020 to 2021 and involves 39 companies within the technology sector. The findings reveal significant fluctuations in the companies' correlations each year. Moreover, the most influential companies vary according to the period, demonstrating position changes. The fact that Inari Amertron (INAR), Revenue Group (REVE), and My EG Services (MYEG) consistently become leading players in different years shows how strong and influential they are in the technology sector. This suggests that these companies have the ability to shape market trends and possibly even control the direction of the industry. The results provide valuable insights for investors and policymakers, aiding in visualizing company relationships and facilitating strategic portfolio optimization.

Analyzing the Relationship between Real Estate Investments and Portfolio Diversification

Real estate has long been considered an attractive investment option for individuals and institutions seeking to build wealth and diversify their portfolios. Unlike traditional investment vehicles such as stocks and bonds, real estate offers unique characteristics that can potentially enhance returns and mitigate risk. This analysis aims to explore the role of real estate investments in portfolio diversification and assess their potential impact on overall portfolio performance. Portfolio diversification is a fundamental principle in investment management, as it helps to spread risk across different asset classes and mitigate the impact of market fluctuations on a portfolio's overall value. By including assets with low or negative correlations, investors can reduce the volatility of their portfolios and potentially achieve higher risk-adjusted returns. Real estate investments, including direct property ownership, real estate investment trusts (REITs), and other real estate-related securities, have traditionally exhibited low correlations with other asset classes, such as equities and bonds. This low correlation can be attributed to the unique characteristics of real estate, including its tangible nature, the presence of rental income streams, and the potential for capital appreciation. Furthermore, real estate investments can provide a hedge against inflation, as property values and rental rates tend to increase during periods of rising prices. This feature makes real estate an attractive diversification option, particularly for investors seeking to protect their portfolios from the eroding effects of inflation. This analysis will delve into the historical performance of real estate investments, examine their risk and return characteristics, and evaluate their potential contribution to portfolio diversification. By examining empirical data and leveraging portfolio optimization techniques, we aim to provide insights into the optimal allocation of real estate investments within a diversified portfolio.

Optimizing Enterprise Marketing Project Portfolios Using Fuzzy Ant Colony Optimization

Enterprise marketing project portfolio optimization is important for business competitiveness, but involves multiple uncertain factors. An integrated model using fuzzy rules was proposed in this paper to enhance ant colony optimization. The fuzzy ant colony algorithm effectively handled ambiguity in project costs, returns, and risks when selecting an optimal portfolio of marketing initiatives. Experiments demonstrated the algorithm efficiency in converging towards high-quality solutions. Case studies indicated the model helped boost customer loyalty and profits through tailored marketing strategies, outperforming conventional approaches. The fuzzy optimization provides an effective decision-making framework for enterprises to maximize marketing effectiveness.

Optimal Portfolio Analysis of Listed Companies in IDX 30

Investors face the dual considerations of return and risk when making investment decisions. Therefore, proper analysis is crucial, especially during the COVID-19 crisis, to achieve maximum returns while minimizing risk. This research used three portfolio optimization models, the Mean-Variance Model, the Mean-Absolute Deviation Model and the Value-at-Risk Model, to construct a stock portfolio. The findings indicated that the Mean-Variance Model can yield an expected return of 16.55% and a portfolio risk of 258.66%. The result from the Mean-Absolute Deviation Model was that the target return is 16%, along with a portfolio risk of 282.43%. Keywords: Portfolio Optimization, Mean-Variance, Mean Absolute Deviation, Value at Risk, R Language, IDX30

Learning to Optimize Contextually Constrained Problems for Real-Time Decision Generation

The topic of learning to solve optimization problems has received interest from both the operations research and machine learning communities. In this paper, we combine ideas from both fields to address the problem of learning to generate decisions to instances of optimization problems with potentially nonlinear or nonconvex constraints where the feasible set varies with contextual features. We propose a novel framework for training a generative model to produce provably optimal decisions by combining interior point methods and adversarial learning, which we further embed within an iterative data generation algorithm. To this end, we first train a classifier to learn feasibility and then train the generative model to produce optimal decisions to an optimization problem using the classifier as a regularizer. We prove that decisions generated by our model satisfy in-sample and out-of-sample optimality guarantees. Furthermore, the learning models are embedded in an active learning loop in which synthetic instances are iteratively added to the training data; this allows us to progressively generate provably tighter optimal decisions. We investigate case studies in portfolio optimization and personalized treatment design, demonstrating that our approach yields advantages over predict-then-optimize and supervised deep learning techniques, respectively. In particular, our framework is more robust to parameter estimation error compared with the predict-then-optimize paradigm and can better adapt to domain shift as compared with supervised learning models. This paper was accepted by Chung Piaw Teo, optimization. Funding: This work was supported in part by the Natural Sciences and Engineering Research Council of Canada. Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/mnsc.2020.03565 .

THE ROLE OF ARTIFICIAL INTELLIGENCE IN FINANCIAL DECISION-MAKING

In the realm of finance, decision-making processes have always been pivotal in determining the success or failure of individuals, businesses, and economies at large. With the advent of Artificial Intelligence (AI), this landscape is undergoing a profound transformation. The integration of AI technologies in financial decision-making is not merely an augmentation of existing processes but a fundamental shift in how decisions are conceived, analysed, and executed. This introduction seeks to delve into the evolving role of AI in financial decision-making, particularly within the context of the Indian financial sector. The rapid pace of technological advancement has propelled AI into the forefront of various industries, and finance is no exception. AI encompasses a spectrum of technologies, including machine learning, natural language processing, deep learning, and predictive analytics, among others. These tools offer the ability to process vast amounts of data, identify patterns, and generate insights at speeds and scales that were previously unimaginable. In the context of financial decision-making, this translates into enhanced capabilities for risk assessment, portfolio optimization, fraud detection, and customer service, among others. India, with its burgeoning economy and vibrant financial markets, stands at the cusp of this AI-driven revolution. The country's financial ecosystem, characterized by a diverse array of institutions ranging from traditional banks to fintech startups, is ripe for AI-led innovation. The adoption of AI in Indian finance is driven by several factors, including the proliferation of digital technologies, the increasing availability of data, and the imperative for efficiency and competitiveness in a globalized economy. Furthermore, initiatives such as the government's Digital India campaign and regulatory efforts to promote fintech innovation have created a conducive environment for AI adoption in the financial sector.

Abstract PO3-09-10: Minority participation in breast cancer therapeutic trials at an NCI-CCC affiliated safety-net system: Advancing equity in clinical research

Abstract Authors: Nasir Qureshi, MBBS Fabian Robles, MS Mai Badran, MBBS Nisha Unni, MD Hsiao C Li, MD Samira Syed, MD Glenda Delgado-Ramos, MD Sangeetha Reddy, MD Heather McArthur, MD Navid Sadeghi, MD Title: Minority participation in breast cancer therapeutic trials at an NCI-CCC affiliated safety-net system: Advancing equity in clinical research. Introduction: Participation of under-represented minorities (URMs) in clinical trials enhances the generalizability of the findings and can also improve clinical outcomes. Multiple factors, such as access, language barrier, distrust, and misconceptions, negatively impact URM enrollments in cancer clinical trials. Here we describe our approach to increase participation of breast cancer URMs in therapeutic trials at an academic-affiliated safety-net system. Methods: Parkland Health (PH) is a large safety-net system in Dallas County, Texas, and is affiliated with the NCI-CCC designated Harold C. Simmons Comprehensive Cancer Center (SCCC) at UT Southwestern. A dedicated team of SCCC research coordinators support the clinical trial operations at PH. In 2020, several steps were taken to increase access and enrollment of URMs in breast cancer trials at PH including review and optimization of the trial portfolio (more trials in curative intent setting), proactive screening, increasing provider awareness, and addition of a bilingual (English/Spanish) patient navigator to the research team. Data for 2018-2019 (pre-intervention) and 2021-2022 (post-intervention) is presented here. Data for 2020 is excluded due to the impact of COVID-19 pandemic on trial operations. Results: Racial and ethnic minorities composed the majority of breast cancer patients at PH (52% Hispanics, 32% Blacks). Ten percent of new patients at PH presented with metastatic disease. In 2018-2019, PH's portfolio included 13 therapeutic trials, 5 of which (38%) were in metastatic setting and 3 required a biomarker. A total of 47 breast cancer patients were enrolled in 11 therapeutic trials at PH, accounting for an average annual accrual rate of 6.1% (7.6% in 2018 and 4.5% in 2019). Accrual to trials in the metastatic setting accounted for 11% (5/47) of enrollments. Two trials in the metastatic setting had zero accruals. In 2021-2022, 16 trials were open at PH, including 7 cooperative group, 4 industry-sponsored (ISTs), and 5 Investigator-Initiated trials (IITs). Five trials (31%) were in metastatic setting and 4 required a genomic biomarker. A total of 87 patients were enrolled in 2021-22, representing an 85% increase in the accruals over the baseline (2018-2019). The annual accrual rates for 2021 and 2022 were 10.5% (43/409) and 10.4% (44/423), respectively. In 2021-22, 63 were enrolled in cooperative group studies, 20 in IITs, and 4 in ISTs. Accrual to trials in the metastatic setting accounted for 6% (5/87) of enrollments. Two studies had zero accruals. The majority of the trial participants (77/87; 89%) belonged to URMs (64 Hispanic, 13 Black). Overall participation rate was 67% (87/129) and was significantly higher in Hispanics compared to Blacks (77% vs 41% - p < 0.001). Among Black patients who declined participation, 42% (8/19) cited lack of interest as the reason, which may reflect distrust or misperceptions regarding clinical trials. Conclusions: The collaboration between academic centers and safety-net systems provides a unique opportunity to promote equity in clinical trials by increasing access and enrollment of URMs. Developing a balanced portfolio that matches the patient population, enhanced screening, and incorporating clinical trial patient navigation at our institution resulted in a sustained increase in the number of URM trial participants. Given the lower participation rate among eligible Black patients, we aim to expand our clinical trial navigation program to better address the distrust and misperceptions regarding clinical trials, as a strategy to further improve trial participation at our institution. Citation Format: Nasir Qureshi, Fabian Robles. Minority participation in breast cancer therapeutic trials at an NCI-CCC affiliated safety-net system: Advancing equity in clinical research [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO3-09-10.

Shocks, Frictions, and Inequality in US Business Cycles

We show how a heterogeneous agent New Keynesian (HANK) model with incomplete markets and portfolio choice can be estimated in state space using a Bayesian approach. To render estimation feasible, the structure of the economy can be exploited and the dimensionality of the model automatically reduced based on the Bayesian priors. We apply this approach to analyze how much inequality matters for the business cycle and vice versa. Even when the model is estimated on aggregate data alone and with a set of shocks and frictions designed to match aggregate data, it broadly reproduces observed US inequality dynamics. (JEL D31, D52, E12, E32, E52, E62)

Extreme time-frequency connectedness across U.S. sector stock and commodity futures markets

This study investigates the extreme time-frequency return connectedness between ten U.S. sectors and commodities from January 2014 to May 2023. Using quantile time-frequency measures, we find the following: Firstly, the total connectedness estimates are more sensitive at shorter frequencies than at longer ones. Secondly, the study reveals varied degrees of contagion during crisis periods. Notably, during COVID-19 (Russia-Ukraine conflict), the contagion is driven by short-term (long-term) shocks, specifically during the bearish (bullish) phase. Thirdly, quantile connectedness measures depict intense correlations around market extremes, underlining dynamic net return-contagion with tailored risk strategies. The shifts in shock transmission roles during bearish and bullish scenarios, along with evolving dynamics across time-frequency horizons, emphasize substantial interconnectedness within the network. Our findings suggest limited diversification scope under extreme market conditions, informing investment decisions, risk management, and portfolio optimization.

Cross validation based transfer learning for cross-sectional non-linear shrinkage: A data-driven approach in portfolio optimization

Enhanced covariance estimation approaches, such as (non-)linear shrinkage, are well established in the literature. Non-linear shrinkage estimators generally minimize a certain loss function regarding statistical assumptions about the future covariance matrix. At the same time, the problem of covariance estimation is traditionally considered from a rather restrictive view since the only available data to determine the estimation parameters is given by the return history of the actual portfolio constituents. In this study, we propose a novel and purely data-driven perspective on covariance estimation. We present a non-linear shrinkage estimator that determines the estimation parameters using cross validation to be historically optimal on a disjoint dataset of assets according to the given objective, such as minimum variance or maximum risk-adjusted return. We then transfer the historically optimal estimation parameters learned on the disjoint dataset to the actual covariance estimation problem. Thereby, the sample eigenvalues are corrected in a purely data-driven way, agnostic to theoretically derived parameters. Another benefit of focusing on disjoint data is that we address the problem of limited data availability in high-dimensional estimation problems when the number of assets exceeds the history length. Our empirical evaluation, based on a total of six stock market indices and various problem dimensions, shows that our approach outperforms existing cross-sectional estimators in minimizing variance and maximizing risk-adjusted return. While our study is limited to the cross-section, the method of parameter selection using cross validation and transfer learning can also be combined with other estimators, such as time-series methods.