Annotated Bibliography for below attached aricles

IEEE TRANSACTIONS ON FUZZY SYSTEMS, VOL. 25, NO. 2, APRIL 2017 245

Guest Editorial Special Issue on Fuzzy Techniques in Financial

Modeling and Simulation

C OMPUTATIONAL intelligence has attracted a signifi-cant and increasing interest from the financial engineering community, and an emerging interest from analytical economics groups. The bar has been raised with the revision of regulations, and the required compliance and risk management. The new rules should be implemented through new processes and sup- ported by developing new computational tools. Computational systems, capturing sentiments, preferences, behavior, and be- liefs, are becoming indispensable in financial applications and desirable in economic analysis. They address problems in the classification of creditworthiness and fraud detection, contribute to the analysis and pricing of financial instruments, and effec- tively support portfolio optimization and investment analysis. They are instrumental in the design of market mechanisms and contagion mechanisms, and are contributing to the simulation of micro- and macro-economic processes.

The armory of fuzzy techniques is capable of addressing chal- lenges encountered in financial engineering and analytical eco- nomics. Fuzzy logic can effectively describe and incorporate experts’ intuition, market participants’ preferences, and eco- nomic agents’ behavior, thus reaching beyond the capabilities of probabilistic models. Fuzzy techniques can also be used in conjunction with the probabilistic models or with other machine learning techniques, such as evolutionary optimization and neu- ral networks, in order to develop effective hybrid approaches for challenges raised in this area. The objective of this special issue is to bring together the most recent advances in the design and application of fuzzy approaches to real problems in financial engineering and analytical economics.

The papers in the special issue have been selected from 72 submissions, following the thorough peer-review process of the IEEE TRANSACTIONS ON FUZZY SYSTEMS. The evaluation pro- cess considered the originality, technical quality, presentation quality, and overall contribution of submitted manuscripts. Six- teen papers were accepted, following three rounds of compre- hensive reviews and revisions. Two of the papers address classi- fication of a broad range of financial data, from creditworthiness to spotting arbitrage opportunities, and further two of the papers focus on stock analysis, price modeling, and volatility forecast- ing. Five papers consider portfolio selection, optimization, and active rebalancing, and another three papers investigate option pricing under reduced assumption. The final four papers extend the analysis toward market mechanisms, international finance,

Digital Object Identifier 10.1109/TFUZZ.2017.2682542

macroeconomic investment analysis, and microeconomic cor- porate management.

The paper titled “Multi-objective evolutionary optimization of type-2 fuzzy rule-based systems for financial data classi- fication” adopts type-2 fuzzy rule-based classifiers generated from data through a multiobjective evolutionary algorithm. The authors develop an effective approach for classifying unbal- anced datasets, compare its performance with the existing meth- ods, and apply it to different financial datasets. The approach successfully classifies creditworthiness for authorizing and for increasing individual and business loans and credit card lim- its, identifies arbitrage opportunities, and detects fraud. Next, the paper titled “Rough information set and its applications in decision-making” combines the concepts of information set and rough set to introduce a structure termed as rough information set (RIS). This structure allows representing both vagueness and imprecision of data and its building blocks are “information granules”—information sets based on fuzzy equivalence rela- tions. The information granules are further casted in the rough set formalism to yield RIS that, in comparison fuzzy rough sets, increase interpretability and the ability to obtain shorter sub- sets of the dataset while maintaining or improving classification accuracy. The approach is applied successfully to credit score analysis.

The paper titled “Modeling stock price dynamics with fuzzy opinion networks” models communications through social net- works among stock investors, and explores the effect of changes in investors’ social network influence on stock price dynamics. An investor’s opinion is represented as a Gaussian fuzzy set, where the uncertainty is based on local, global or external refer- ence schemes in order to model different scenarios of forming opinion. The analysis and empirical results prove that investors converge to different groups in finite time under the local refer- ence scheme. On the other hand, they converge to a consensus within finite time under the global or external reference schemes, where the consensus may drift in time under the external refer- ence case. Next, the paper titled “Evolving possibilistic fuzzy modeling for realized volatility forecasting with jumps” sug- gests an approach to improve forecasting performance, based on an extension of the possibilistic fuzzy c-means clustering and on using the functional fuzzy rule-based models. The evolving na- ture of the model allows adding or removing clusters, where the update is based on statistical distance-like criteria dictated by input data. This evolving possibilistic fuzzy approach is highly efficient in modeling realized volatility in terms of forecasting

1063-6706 © 2017 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications standards/publications/rights/index.html for more information.

246 IEEE TRANSACTIONS ON FUZZY SYSTEMS, VOL. 25, NO. 2, APRIL 2017

accuracy and improves robustness to noise and outliers in data, as confirmed with the empirical analysis based on data for major equity market indexes.

The paper titled “FN-TOPSIS: Fuzzy networks for rank- ing traded equities” extends the technique for ordering of preference by similarity to ideal solution (TOPSIS) and uses networked rule bases. The FN-TOPSIS approach captures both imprecisions inherent in financial data and in the decision- making processes on them, in order to solve multicriteria decision-making problems where benefit and cost criteria are presented as subsystems. This allows the decision maker to evaluate the performance of each alternative equity in portfolio selection, while observing the performance for both benefit and cost criteria. The new approach significantly improves trans- parency and accuracy in comparison with the existing TOPSIS methods, as proved by the empirical results. Next, the paper titled “Stock picking by probability-possibility approaches” proposes probabilistic and possibilistic fusion operators, in order to merge several technical indicators and to improve stock selection. The performance of portfolios selected with the new approach is evaluated through cumulative returns and Sharpe ratios, and it is shown that the possibilistic framework is more robust to redundant sources of information, in comparison with the probabilistic framework. The empirical results reveal the potential of technical indicators fusion for improving portfolio selection, and confirm effect of fusion parameters. Further, the main idea in the paper titled “Mean-variance portfolio selection with the ordered weighted average” is to replace the classical mean–variance approach with the ordered weighted average operator. This allows for considering different degrees of decision makers’ optimism and pessimism during the selection process, and suggests a framework for dealing with the attitudinal tendencies of decision makers. The study concludes with empirical analysis. The next paper on portfolios analysis is titled “Adaptive budget-portfolio investment optimization under risk tolerance ambiguity” and develops a two-stage adaptive optimization model to capture decision dynamics driven by risk tolerance ambiguity. The authors introduce the concept of risk- neutral budget threshold and represent it as a fuzzy set granule, in order to construct the ambiguous risk tolerance curve and capture realistically decision-makers’ risk aversion attitudes. The model is further transformed into a mixed integer linear programing and the Benders decomposition is implemented to enhance the scalability. The empirical analysis is based on securities traded on the New York Stock Exchange. The final paper on portfolio optimization is titled “Fuzzy decision theory based metaheuristic portfolio optimization and active rebalancing using interval type-2 fuzzy sets,” and considers a problem including the three tasks of portfolio optimization, market scenario forecasting, and portfolio rebalancing. The first phase depends on metaheuristics to arrive at the optimal portfolio, the second phase adopts a Monte Carlo simulation to generate future market scenarios, and the third phase employs interval type-2 fuzzy sets to exploit the uncertainty of generated scenarios and arrive at the optimal rebalanced portfolio. The empirical analysis is based on high-risk constituents of the BSE 200 index, and includes comparative results using other algorithms.

The first paper on options is titled “Fuzzy approaches to op- tion price modelling” and presents a comprehensive review of studies in this area, differentiating between discrete-time and continuous-time models, and paying special attention to real options. The review addresses both direct and inverse problems in option pricing, where in the former, the stochastic process for the underlying asset is assumed and the option prices are derived, while in the latter, the option prices are given and used to infer the underlying asset process. An emphasis is placed on the need of empirical analysis to assess improvements in the modeling of imprecise data with fuzzy sets and fuzzy random variables. Next, the paper titled “Option pricing with application of Levy processes and the minimal variance equivalent martin- gale measure under uncertainty” considers European options, and applies stochastic analysis and fuzzy sets theory. Option valuation is obtained by using the minimal variance equiva- lent martingale measure, Levy transformation of characteristic triplets, and fuzzified model parameters. This valuation allows capturing various market uncertainties, and is applied in the pro- posed method for automatized decision making. Finally, numer- ical examples illustrate the theoretical results. The third paper on options is titled “Quanto European option pricing with am- biguous return rates and volatilities” and implements set-valued stochastic differential inclusions to describe the Black–Scholes quanto model. Nonlinear minimum and maximum conditional expectations are used to obtain the upper and lower bounds of the contingent claim. The conditional expectations are calcu- lated with backward stochastic differential equations, and the effectiveness of the model is demonstrated with a numerical example.

The last group of papers addresses problems of a broader scope, crossing from the financial into the economics domain. The first paper is titled “A comparison of bidding strategies for online auctions using fuzzy reasoning and negotiation decision functions,” and designs bidding strategies aiming to forecast the bid amounts for buyers at a point in time. This is based on their bidding behavior and their valuation of auctioned items, and addresses the issues of bid timing and auction selection among simultaneous auctions for similar items. The bidding strategies are designed implementing the two technical approaches—a Mamdani method with regression analysis and negotiation de- cision functions—and the experimental results show that agents following these strategies outperform other agents in terms of success rate and expected utility. Next, the paper titled “Fuzzy dynamical system scenario simulation-based cross-border finan- cial contagion analysis: A perspective from international capi- tal flows” analyzes the volatility of international capital flows and contagion effects, for a panel of 50 countries in emerg- ing markets and advanced economies from 1980 to 2011. The comovement of the source country of financial turbulence and the volatility-affected country is described as a fuzzy dynamical system with coupled driving and response systems, where the uncertain strength of coupling is modeled with interval Type-2 fuzzy sets. The model explains different volatility transmission patterns and extends to include macroeconomic control, provid- ing insights for policymakers. The empirical analysis is based on the 2008 global financial crisis. Furthermore, the paper ti- tled “Multiobjective investment policy for nonlinear stochastic

IEEE TRANSACTIONS ON FUZZY SYSTEMS, VOL. 25, NO. 2, APRIL 2017 247

financial system: Fuzzy approach” proposes multiobjective fuzzy investment to achieve Pareto-minimum investment cost and risk at the macrolevel, where the investment model includes as variables the interest rate, investment demand, and price in- dex. A Takagi–Sugeno fuzzy model is used to transform this multiobjective optimization problem (MOP) with Hamilton– Jacobi inequalities into a linear matrix inequalities-constrained MOP. A multiobjective evolutionary algorithm is developed to efficiently identify Pareto optimal solutions for the latter prob- lem, where investment policies achieve desired minimum invest- ment cost and match different risk preferences. The efficiency of the approach is demonstrated with a numerical example mimick- ing emerging markets. Finally, the paper titled “A fuzzy control model for restraint of bullwhip effect in uncertain closed-loop supply chain with hybrid recycling channels” addresses a prob- lem of corporate management at microeconomic level. The ba- sic models for a closed-loop supply chain are converted into a nonlinear fuzzy switching model based on a discrete Takagi– Sugeno fuzzy control system. A fuzzy robust control method is further utilized to reduce the impacts caused by internal and external uncertain factors on the closed-loop supply chain. The empirical simulations confirm the feasibility and effectiveness of the constructed fuzzy control system.

ACKNOWLEDGMENT

We would like to thank the authors for sharing their recent research ideas and for their dedication to this applied research area, and the numerous reviewers for their constructive com- ments on improving the quality of submitted manuscripts. We sincerely thank Prof. C.-T. Lin, Editor-in-Chief of the IEEE

TRANSACTIONS ON FUZZY SYSTEMS, for accepting the proposal for this special issue and for his constant and effective support throughout the two years of putting the issue together. We hope that the readers will enjoy this special issue and find it stimu- lating and thought-provoking, and that it will inspire others to join the community of researchers in computational finance and economics.

A. SERGUIEVA, Guest Editor Research Centre for Blockchain Technologies Department of Computer Science University College London London WC1E 6EA, U.K.

H. ISHIBUCHI, Guest Editor Department of Computer Science and Engineering Southern University of Science and Technology Shenzhen 518055, China Department of Computer Science and Intelligent Systems Osaka Prefecture University Sakai 599-8531, Japan

R. R. YAGER, Guest Editor Machine Intelligence Institute Iona College New Rochelle, NY 10805 USA

V. P ALADE, Guest Editor Department of Computing Coventry University Coventry CV1 5FB, U.K.

Antoaneta Serguieva (M’02–SM’11) received the M.Sc. degree in systems engineering from the Technical University of Sofia, Sofia, Bulgaria, in 1989; the MBA in finance from the University of National and World Economy, Sofia, Bulgaria, in 1997; and the Ph.D. degree in computer science from Brunel University, London, U.K., in 2004.

In 2017, she became a Research Associate with the Systemic Risk Centre, London School of Economics, since March 2016, she has been a member in the Research Centre for Blockchain Technologies, University College London (UCL), London, U.K., and since 2010, has been in the Department of Computer Science, UCL. Since 2014, she has been a Visiting Research Fellow at the Centre for Computational Finance and Economic Agents, University of Essex, Colchester, U.K. She has recently completed a fellowship in advanced analytics at the Bank of England.

Dr. Serguieva Chaired the Computational Finance and Economics Technical Committee of the IEEE Computational Intelligence Society (2014–2015), and Chaired the 2014 IEEE Conference on Computational Intelligence for Financial Engineering and Economics.

Hisao Ishibuchi (M’93–SM’10–F’14) received the B.S. and M.S. degrees in precision mechanics from Kyoto University, Kyoto, Japan, in 1985 and 1987, respectively, and the Ph.D. degree in computer science from Osaka Prefecture University, Sakai, Japan, in 1992.

Since 1987, he has been with Osaka Prefecture University. Since 2017, he has been a Chair Professor at the Southern University of Science and Technology, Shenzhen, China. His research interests include fuzzy rule-based classifiers, evolutionary multiobjective and many-objective optimization, memetic algorithms, and evolutionary games.

Dr. Ishibuchi was the Vice-President for Technical Activities of the IEEE Computational Intelligence Society (CIS) from 2010 to 2013. He is currently an AdCom member of the IEEE CIS (2014–2019), an IEEE CIS Distinguished Lecturer (2015–2017), and the Editor-in-Chief of the IEEE COMPUTATIONAL INTELLIGENCE MAGAZINE (2014–2017).

248 IEEE TRANSACTIONS ON FUZZY SYSTEMS, VOL. 25, NO. 2, APRIL 2017

Ronald R. Yager (LF’11) received the undergraduate degree from the City College of New York, New York, NY, USA, and the Ph.D. degree from the Polytechnic University of New York, Brooklyn, NY.

He is currently the Director of the Machine Intelligence Institute and a Professor of Information Systems at Iona College, New Rochelle, NY. He has worked in the area of machine intelligence for more than 25 years. He served at the National Science Foundation as a Program Director in the Information Sciences Program. He was a NASA/Stanford Visiting Fellow and a Research Associate at the University of California, Berkeley, CA, USA. He has been a Lecturer at NATO Advanced Study Institutes. He is a Distinguished Honorary Professor at the Aalborg University, Aalborg, Denmark and a Distinguished Visiting Scientist at King Saud University, Riyadh, Saudi Arabia. He has published more than 500 papers and 15 books in areas related to fuzzy sets, decision making under uncertainty, and the fusion of information. He is among the world’s top 1% most highly cited researchers with more than 7000 citations.

Dr. Yager is a Fellow of the New York Academy of Sciences and the Fuzzy Systems Association. He is the Editor-in-Chief of the International Journal of Intelligent Systems and serves on the editorial board of numerous technology journals. He received the Lifetime Achievement Award by the Polish Academy of Sciences for his contributions and the IEEE Computational Intelligence Society Pioneer Award in fuzzy systems.

Vasile Palade joined the Department of Computing, Coventry University, Coventry, U.K., in September 2013 after working for many years with the Department of Computer Science, Uni- versity of Oxford, Oxford, U.K. He is the author of more than 120 papers in journals and conference proceedings as well as books on machine learning/computational intelligence and applications. His research interests include machine learning with various applications.

He is a member of the IEEE Computational Intelligence Society. He has delivered keynote addresses and chaired international conferences on machine learning and applications. He is an Associate Editor for several reputed journals.