An Adaptive Algorithm for Log-Optimal Portfolio and Its Theoretical Analysis

ZX Ye,JG Huang

DOI: https://doi.org/10.1109/isit.2000.866531

2001-01-01

Abstract:An adaptively random searching algorithm is proposed for log-optimal portfolio. Under reasonable conditions the convergence of this algorithm is shown. The numerical results using this algorithm for real data from Shanghai Stock Exchanges are satisfactory

What problem does this paper attempt to address?

An Riemannian Geometry Underlying Stochastic Algorithm for the Risk-free Log-Optimal Portfolio Problem

董承非,黄建国

DOI: https://doi.org/10.3969/j.issn.1006-3080.2004.01.024

2004-01-01

Abstract:The risk-free log-optimal portfolio problem is very valuable in computational finance in practice. This paper proposes an intrinsic natural gradient-based stochastic algorithm for solving the pro-(blem). The advantage of the algorithm is that the resulting iteration vector always satisfies the constrained conditions, and it is also an interior point algorithm in essence, so has a good rate of convergence. Finally, this algorithm is applied to the real data from the Exchange Institute of Shanghai Security. The numerical results are satisfactory and the reasonable financial explanation is also suggested.
An additive on-line portfolio selection algorithm

Lan-Jun Lao

DOI: https://doi.org/10.1080/716067188

2002-01-01

Abstract:An on-line portfolio management algorithm using additive update rule is presented with consideration of proportional transaction costs. After establishing a relationship between transaction costs and the distance of two portfolio vectors, we solve an optimization problem making a balance between high wealth growth rate and low transaction costs. It is shown that the presented portfolio strategy is universal relative to the set of all buy-and-hold portfolios in the sense that it achieves the same asymptotic exponential growth rate of wealth as the best buy-and-hold portfolio determined with hindsight. The simulation examples using the real data from the London Stock Exchange show the efficiency and the computational simplicity of the algorithm.
The Adaptive Genetic Algorithms for Portfolio Selection Problem

Wei-Guo Zhang,Wei CHEN,Ying-Luo Wang

2006-01-01

Abstract:Summary Genetic algorithms (GA) are stochastic search techniques based on the mechanics of natural selection and natural genetics. In this paper, the adaptive genetic algorithms are applied to solve the portfolio selection problem in which there exist both probability constraint on the lowest return rate of portfolio and lower and upper bounds constraints on the investment rates to assets. First, the stochastic model of portfolio selection and it's the reliability decision are presented. Second, the adaptive genetic algorithm to solve the reliability decision is given. Finally, a numerical example of portfolio selection problem is given to illustrate our proposed effective means.
A Study on Algorithm for Portfolio Based on the Strategy of Equal-Risk

GAO Jian-xi,DONG Hong-guang,LIU Yuan-yuan,LIU jian-guo

DOI: https://doi.org/10.3969/j.issn.1673-9965.2005.05.005

2005-01-01

Abstract:A new algorithm based on the strategy of equal-risk,is presented in the present paper.The algorithm adopts strategy of step-by-step adjustment,and especially considering the smallest exchange fee,utilize the non-risk bank investment to do corresponding adjustments.At last the relationship between the lowest risk and the corresponding highest profit has been found.The algorithm overcomes the problem that the result is sensitive to the scale of investment,so it is a universal algorithm not relative to the scale of investment.The numerical results of the study suggest that the algorithm is very effective.
Robust and adaptive algorithms for online portfolio selection

Theodoros Tsagaris,Ajay Jasra,Niall Adams

DOI: https://doi.org/10.1080/14697688.2012.691175

IF: 1.986

2012-11-01

Quantitative Finance

Abstract:We present an online approach to portfolio selection. The motivation is within the context of algorithmic trading, which demands fast and recursive updates of portfolio allocations as new data arrives. In particular, we look at two online algorithms: Robust-Exponentially Weighted Least Squares (R-EWRLS) and a regularized Online minimum Variance algorithm (O-VAR). Our methods use simple ideas from signal processing and statistics, which are sometimes overlooked in the empirical financial literature. The two approaches are evaluated against benchmark allocation techniques using four real data sets. Our methods outperform the benchmark allocation techniques in these data sets in terms of both computational demand and financial performance.

economics,social sciences, mathematical methods,mathematics, interdisciplinary applications,business, finance
Log-Optimal Portfolio Selection Using the Blackwell Approachability Theorem

Vladimir V'yugin

DOI: https://doi.org/10.48550/arXiv.1410.5996

2015-06-28

Abstract:We present a method for constructing the log-optimal portfolio using the well-calibrated forecasts of market values. Dawid's notion of calibration and the Blackwell approachability theorem are used for computing well-calibrated forecasts. We select a portfolio using this "artificial" probability distribution of market values. Our portfolio performs asymptotically at least as well as any stationary portfolio that redistributes the investment at each round using a continuous function of side information. Unlike in classical mathematical finance theory, no stochastic assumptions are made about market values.

Artificial Intelligence,Portfolio Management
An Evolutionary Algorithm with a New Operator and an Adaptive Strategy for Large-Scale Portfolio Problems

Yi Chen,Aimin Zhou,Liang Dou

DOI: https://doi.org/10.1145/3205651.3205712

2018-01-01

Abstract:A portfolio optimization problem involves optimal allocation of finite capital to a series of assets to achieve an acceptable trade-off between profit and risk in a given investment period. In the paper, the extended Markowitz's mean-variance portfolio optimization model is studied with some practical constraints. We introduce a new operator and an adaptive strategy for improving the performance of the multi-dimensional mapping algorithm (MDM) proposed specially for the portfolio optimization. Experimental results show that the modification is efficient on tackling large-scale portfolio problems.
A stock selection algorithm hybridizing grey wolf optimizer and support vector regression

Meng Liu,Kaiping Luo,Junhuan Zhang,Shengli Chen

DOI: https://doi.org/10.1016/j.eswa.2021.115078

IF: 8.5

2021-10-01

Expert Systems with Applications

Abstract:<p>Artificial intelligence remarkably facilitates quantitative investment. A latest intelligent search algorithm, grey wolf optimizer, is well integrated with support vector regression machine to obtain the optimal portfolio. The performance of the hybrid algorithm is empirically investigated through transactional and financial data from stock markets of America and China. The experimental results indicate that (i) the proposed algorithm is able to stably achieve excess returns; (ii) compared with genetic algorithm, particle swarm optimization, gravitational search algorithm and harmony search, the enhanced grey wolf optimizer significantly boots the predictive performance of support vector regression machine; (iii) the proposed algorithm can achieve the better profitability and the higher reliability in Chinese A-share market.</p>

computer science, artificial intelligence,engineering, electrical & electronic,operations research & management science
An Empirical Study on Discrete Optimization Models for Portfolio Selection

Xueting Cui,Xiaoling Sun,Dan Sha

DOI: https://doi.org/10.3934/jimo.2009.5.33

2009-01-01

Journal of Industrial and Management Optimization

Abstract:In this paper, we investigate four discrete optimization models arising from single period portfolio selection: Mean-variance model, mean-absolute-deviation model, minimax model and conditional Value-at-Risk model. These four models are established by considering the minimal transaction unit and the cardinality constraint in real-world investment practice. Extensive computational results are reported to compare the features of the models. We evaluate the performance of the models by analyzing the in-sample and out-of-sample numerical results with real data from Shanghai Stock Exchange.
Binary Switch Portfolio

Tengfei Li,Kani Chen,Yang Feng,Zhiliang Ying

DOI: https://doi.org/10.1080/14697688.2016.1223337

IF: 1.986

2016-01-01

Quantitative Finance

Abstract:We propose herein a new portfolio selection method that switches between two distinct asset allocation strategies. An important component is a carefully designed adaptive switching rule, which is based on a machine learning algorithm. It is shown that using this adaptive switching strategy, the combined wealth of the new approach is a weighted average of that of the successive constant rebalanced portfolio and that of the 1/N portfolio. In particular, it is asymptotically superior to the 1/N portfolio under mild conditions in the long run. Applications to real data show that both the returns and the Sharpe ratios of the proposed binary switch portfolio are the best among several popular competing methods over varying time horizons and stock pools.
Research on Stock Market Portfolio Optimization Using Stochastic Matrix Theory and Genetic Algorithm

YanJin Zhang

DOI: https://doi.org/10.1155/2022/8355934

IF: 1.43

2022-08-26

Mathematical Problems in Engineering

Abstract:Financial engineering is a synthesis of modern finance, information technology, engineering, and other fields. Under the complex and volatile market conditions, how to allocate assets reasonably to minimize the benefits and risks assumed is a very important topic. Using the principle of random matrix, the correlation statistical characteristics of stocks are analyzed, the noise reduction and reconstruction algorithm of the correlation matrix is discussed, and a new noise reduction algorithm, two-point determination method, is given. The investment portfolio is numerically simulated, and its parameters are set by the orthogonal experimental method. Through analysis, it is found that the eigenvector contained in the maximum eigenvalue of the correlation matrix can be regarded as a market indicator. The "two-point determination method" has a good effect in "denoising" and can provide operational decision-making and technical support for portfolios.

engineering, multidisciplinary,mathematics, interdisciplinary applications
Efficient Optimisation for Portfolio Selections under Prospect Theory

Chunhui Xu,Chao Gong,Xiangming Xi,Shuning Wang

DOI: https://doi.org/10.1504/ajmsa.2018.10016649

2018-01-01

Asian J of Management Science and Applications

Abstract:The prospect theory (PT) is one of the most useful tools for portfolio optimisation. The main concept of PT is to use a S-shaped value function to depict how human beings' mental behaviour affect their investment decisions under different risk levels. However, the complexity in the theory results in the difficulty in the proposal of efficient algorithms for global optimisation. In order to make improvements, we first approximate the S-shaped value function in PT with a piecewise linear (PWL) surrogate model, and equivalently transform the resulted problem into a continuous concave piecewise linear maximisation problem. Despite of the non-smoothness and non-convexity of the problem, we propose two local search algorithms based on the interior point method, and present the theoretical analysis on the convergence. Moreover, we propose a global search algorithm based on the proposed local search algorithms and the γ valid cut method in concave optimisation. The numerical experiments on the historical data of different assets obtained from Yahoo concern the comparisons of the proposed algorithms and the existing methods in the literature. The results confirm the performances of the proposed algorithm on efficiency and accuracy.
Application and Comparative Study of Optimization Algorithms in Financial Investment Portfolio Problems

Shuai Liu,Chenglin Xiao

DOI: https://doi.org/10.1155/2021/3462715

2021-11-28

Mobile Information Systems

Abstract:Portfolio theory mainly studies how to optimize the allocation of assets under the premise of maximizing expected returns and minimizing investment risks. In view of the instability of the financial market, a diversified investment portfolio can help control the loss of the investment portfolio. In addition to paying attention to the safety and return of asset allocation, we cannot ignore the liquidity of assets, that is, their liquidity. Adding high-liquidity products to asset allocation, such as equity investment, can better control the financial cash flow in response to emergencies. One of the ways to make assets flow is to securitize assets and sell them to the market. In order to revitalize the stock assets, good investment efficiency is a necessary choice for financial investment. Various financial products and their derivatives continue to enter people’s vision. There are many financial products in reality, and optimizing the investment portfolio can bring high economic benefits. The purpose of this paper is to study the application of optimization algorithms in financial portfolio problems. (1) Monetary policy remains prudent and neutral. It is not easy to expect flooding, but flexibility is required in complex situations. (2) Financial resources are tilted towards innovation and transformation and capital markets, which is beneficial to the development of capital markets in the medium and long term. (3) Unblocking the transmission mechanism is conducive to lenient credit and tapping the wrong killing opportunities in private enterprise debt. (4) Banks and other financial institutions have moderate pressure to give benefits to entities, but in the long run, the interests of the two are consistent. (5) Finance risk prevention will continue, orderly breaking the rigid exchange and reshaping the financial structure and ecology. (6) The pace of opening up of the financial industry has accelerated, and the bond market investor structure has improved. In this paper, we establish different optimization schemes to compare and study the portfolio problem and then use MATLAB to solve the modeling and programming problem, calculate the highest return rate and the lowest risk value before and after optimization, and then make a comparative analysis to get a better optimization scheme. The results show that the genetic algorithm model is superior to the quadratic programming method in terms of risk control. The minimum risk of portfolio optimization through genetic algorithm has been reduced by about 40%, and the maximum return has increased by about 25%. The comprehensive optimization effect is better than the quadratic planning method and ultimately can obtain higher economic benefits. It can be seen that the optimization algorithm is of great significance for the comparative study of financial portfolio problems.

computer science, information systems,telecommunications
On Solving Robust Log-Optimal Portfolio: A Supporting Hyperplane Approximation Approach

Chung-Han Hsieh

DOI: https://doi.org/10.1016/j.ejor.2023.09.040

2022-02-08

Abstract:A {log-optimal} portfolio is any portfolio that maximizes the expected logarithmic growth (ELG) of an investor's wealth. This maximization problem typically assumes that the information of the true distribution of returns is known to the trader in advance. However, in practice, the return distributions are indeed {ambiguous}; i.e., the true distribution is unknown to the trader or it is partially known at best. To this end, a {distributional robust log-optimal portfolio problem} formulation arises naturally. While the problem formulation takes into account the ambiguity on return distributions, the problem needs not to be tractable in general. To address this, in this paper, we propose a {supporting hyperplane approximation} approach that allows us to reformulate a class of distributional robust log-optimal portfolio problems into a linear program, which can be solved very efficiently. Our framework is flexible enough to allow {transaction costs}, {leverage and shorting}, {survival trades}, and {diversification considerations}. In addition, given an acceptable approximation error, an efficient algorithm for rapidly calculating the optimal number of hyperplanes is provided. Some empirical studies using historical stock price data are also provided to support our theory.

Optimization and Control,Computational Finance,Portfolio Management,Risk Management
A hybrid intelligent algorithm for portfolio selection problem with fuzzy returns

Xiang Li,Yang Zhang,Hau-San Wong,Zhongfeng Qin

DOI: https://doi.org/10.1016/j.cam.2009.07.019

IF: 2.872

2009-11-01

Journal of Computational and Applied Mathematics

Abstract:Portfolio selection theory with fuzzy returns has been well developed and widely applied. Within the framework of credibility theory, several fuzzy portfolio selection models have been proposed such as mean–variance model, entropy optimization model, chance constrained programming model and so on. In order to solve these nonlinear optimization models, a hybrid intelligent algorithm is designed by integrating simulated annealing algorithm, neural network and fuzzy simulation techniques, where the neural network is used to approximate the expected value and variance for fuzzy returns and the fuzzy simulation is used to generate the training data for neural network. Since these models are used to be solved by genetic algorithm, some comparisons between the hybrid intelligent algorithm and genetic algorithm are given in terms of numerical examples, which imply that the hybrid intelligent algorithm is robust and more effective. In particular, it reduces the running time significantly for large size problems.

mathematics, applied
A Minimax Portfolio Selection Strategy with Equilibrium.

XT Deng,ZF Li,SY Wang

DOI: https://doi.org/10.1016/j.ejor.2004.01.040

IF: 6.4

2004-01-01

European Journal of Operational Research

Abstract:A new minimax model on optimal portfolio selection with uncertainty of both randomness and estimation in inputs is established and the corresponding optimal portfolio is derived analytically. Based on this result, a sufficient condition for the existence and uniqueness of a nonnegative equilibrium price system under which the total demand and supply of each asset are equal is provided and an explicit formula for such a price system is obtained. Furthermore, some properties of the equilibrium are discussed.
On admissible efficient portfolio selection: Models and algorithms

Wei-Guo Zhang,Wen-An Liu,Ying-Luo Wang

DOI: https://doi.org/10.1016/j.amc.2005.09.085

IF: 4.397

2006-01-01

Applied Mathematics and Computation

Abstract:The admissible efficient portfolio selection problem for risky assets has been discussed by Zhang and Nie. In this paper, the admissible efficient portfolio model is proposed under general investment constrains on risky assets and a risk-less asset. The closed form solution of the admissible efficient frontiers are derived from three cases: the risk-less asset can either be only lent, or borrowed, or both lent and borrowed. The upper and lower admissible efficient frontiers are developed by the expected return and risk to be estimated optimistically and pessimistically, respectively.
From deterministic to stochastic: an interpretable stochastic model-free reinforcement learning framework for portfolio optimization

Zitao Song,Yining Wang,Pin Qian,Sifan Song,Frans Coenen,Zhengyong Jiang,Jionglong Su

DOI: https://doi.org/10.1007/s10489-022-04217-5

Abstract:As a fundamental problem in algorithmic trading, portfolio optimization aims to maximize the cumulative return by continuously investing in various financial derivatives within a given time period. Recent years have witnessed the transformation from traditional machine learning trading algorithms to reinforcement learning algorithms due to their superior nature of sequential decision making. However, the exponential growth of the imperfect and noisy financial data that is supposedly leveraged by the deterministic strategy in reinforcement learning, makes it increasingly challenging for one to continuously obtain a profitable portfolio. Thus, in this work, we first reconstruct several deterministic and stochastic reinforcement algorithms as benchmarks. On this basis, we introduce a risk-aware reward function to balance the risk and return. Importantly, we propose a novel interpretable stochastic reinforcement learning framework which tailors a stochastic policy parameterized by Gaussian Mixtures and a distributional critic realized by quantiles for the problem of portfolio optimization. In our experiment, the proposed algorithm demonstrates its superior performance on U.S. market stocks with a 63.1% annual rate of return while at the same time reducing the market value max drawdown by 10% when back-testing during the stock market crash around March 2020.
Robust Log-Optimal Strategy with Reinforcement Learning

Yifeng Guo,Xingyu Fu,Yuyan Shi,Mingwen Liu

DOI: https://doi.org/10.48550/arXiv.1805.00205

2018-05-01

Abstract:We proposed a new Portfolio Management method termed as Robust Log-Optimal Strategy (RLOS), which ameliorates the General Log-Optimal Strategy (GLOS) by approximating the traditional objective function with quadratic Taylor expansion. It avoids GLOS's complex CDF estimation process,hence resists the "Butterfly Effect" caused by estimation error. Besides,RLOS retains GLOS's profitability and the optimization problem involved in RLOS is computationally far more practical compared to GLOS. Further, we combine RLOS with Reinforcement Learning (RL) and propose the so-called Robust Log-Optimal Strategy with Reinforcement Learning (RLOSRL), where the RL agent receives the analyzed results from RLOS and observes the trading environment to make comprehensive investment decisions. The RLOSRL's performance is compared to some traditional strategies on several back tests, where we randomly choose a selection of constituent stocks of the CSI300 index as assets under management and the test results validate its profitability and stability.

Portfolio Management

An Adaptive Algorithm for Log-Optimal Portfolio and Its Theoretical Analysis

An Riemannian Geometry Underlying Stochastic Algorithm for the Risk-free Log-Optimal Portfolio Problem

An additive on-line portfolio selection algorithm

The Adaptive Genetic Algorithms for Portfolio Selection Problem

A Study on Algorithm for Portfolio Based on the Strategy of Equal-Risk

Robust and adaptive algorithms for online portfolio selection

Log-Optimal Portfolio Selection Using the Blackwell Approachability Theorem

An Evolutionary Algorithm with a New Operator and an Adaptive Strategy for Large-Scale Portfolio Problems

A stock selection algorithm hybridizing grey wolf optimizer and support vector regression

An Empirical Study on Discrete Optimization Models for Portfolio Selection

Binary Switch Portfolio

Research on Stock Market Portfolio Optimization Using Stochastic Matrix Theory and Genetic Algorithm

Efficient Optimisation for Portfolio Selections under Prospect Theory

Application and Comparative Study of Optimization Algorithms in Financial Investment Portfolio Problems

On Solving Robust Log-Optimal Portfolio: A Supporting Hyperplane Approximation Approach

A hybrid intelligent algorithm for portfolio selection problem with fuzzy returns

A Minimax Portfolio Selection Strategy with Equilibrium.

On admissible efficient portfolio selection: Models and algorithms

From deterministic to stochastic: an interpretable stochastic model-free reinforcement learning framework for portfolio optimization

Robust Log-Optimal Strategy with Reinforcement Learning