Optimal portfolio selection using rule-based fuzzy expert system

Number of pages: 166 File Format: word File Code: 30698
Year: 2014 University Degree: Master's degree Category: Management
  • Part of the Content
  • Contents & Resources
  • Summary of Optimal portfolio selection using rule-based fuzzy expert system

    Dissertation for Master degree (M.A)

    Trend: Finance

    Dissertation abstract (including summary, objectives, implementation methods and obtained results):

    The purpose of this research is to build suitable portfolios by considering the level of risk tolerance of investors and their preferences in a flexible, practical and realistic way. For this purpose, a rule-based fuzzy expert system is built to support investment managers in their mid-term investment decisions. The performance of the proposed expert system has been investigated by the data of 106 stocks traded in the Tehran Stock Exchange between 2014 and 2016. The performance of the proposed expert system has been analyzed in terms of risk tolerance and the length of the investment period, compared to the market average. The results show that the proposed expert system performs better than the market in most cases. Also, according to our expectations, the performance of this expert system is better for the risk-averse investor in the medium term.

    Statement of the problem

    One of the ways to invest and form a portfolio of assets is investing in the stock exchange. In developed countries, most of the investments are made through financial markets (stock exchanges). Choosing and choosing the stocks of the companies present in the stock exchange and forming the optimal stock portfolio depends on several factors that complicate decision making for analysts and experts. Many researches and studies have been done in the field of determining the priority of stock selection criteria and forming the optimal portfolio according to different criteria and using modern models in interaction with each other. However, quantitative models simultaneously consider fundamental and technical approaches for portfolio construction. The main issue of this research is stock evaluation and portfolio construction using an expert system that considers both approaches. Now the question is whether it is possible to use this system and the selected inputs to choose the portfolio that is the most profitable stock portfolio, and whether this decision will be made more quickly and accurately than the classic models of portfolio selection? 1-2 Research Objectives The main objective of this study is to build the optimal stock portfolio using a rule-based fuzzy expert system. Other goals include: examining the performance of the model in portfolio selection, interaction between decision makers and the high-level system, supporting portfolio managers in investment decisions.

    1-3 The importance of the research topic and the motivation for choosing it

    A) Optimal investment, in addition to increasing the wealth of shareholders, also causes the economic growth of any country in the macro view.      

    B) The use of new and more accurate models can increase the return on investment and also lead to optimal capital allocation.

    C) Entering the capital market and using new tools in order to obtain more efficiency is a step towards making the market more efficient.

    Second hypothesis) The proposed expert system has better performance for risk-averse investors.

    Third hypothesis) The proposed expert system has better performance in the medium term.

    1-5 Research model

    Figure 1-1 Work process

    1-6 Operational definitions of variables and keywords

    Expert system: a computer system including a body of knowledge that It is well organized, it imitates the problem solving skills of experts in a limited skill domain (Bahrammirzaee, 2010). In some parts, the concept of expert system refers only to basic knowledge systems (Rada, 2008).

    A rule-based expert system: a computer program that is able to use knowledge-based information through sets of inferential procedures to solve problems that are difficult enough to require significant human skill to solve them (Harmon P, 1985).

    Fundamental analysis: fundamental analysis (or fundamental) is one of the methods of capital market analysis and price prediction. The fundamental analysis of a company includes the analysis of financial reports and financial health of the company, management and competitive advantages, competitors and relevant markets.

    Technical analysis: It is a method of predicting prices in the market by studying the past state of the market. In this analysis, it is possible to predict the price situation in the future by examining the changes and fluctuations of prices and volume of transactions and supply and demand. This method of analysis is widely used in foreign markets, stock market, gold market and other precious metals. The most important principle of technical analysis is that the price reflects everything about a share (Latfi, Darwish, 2015, 1).

    Technical index: A technical indicator is a tool through which you can analyze price behavior (same source, 2015, 15).

    1-7 research method

    It is a survey type description. Library method is used to collect scientific literature and background check. Also, the current research is carried out within the framework of the basics of modeling and the use of information systems in the field of financial management.

    1-8 scope of research

    a) spatial domain: all companies admitted to the stock exchange

    b) temporal domain: between the years 1384 and 1389

    c) thematic domain: in the field of financial management and capital market

    1-9 Population and sample size

    The statistical population of this research includes companies active in the Tehran Stock Exchange between 1384 and 1389.

    From the statistical population and based on the following criteria, 106 companies were selected as a sample:

    The companies were admitted to the stock market by the end of March 1383 and were delisted from the stock market by 1389.

    Fiscal year of all companies must be the end of March.

    The company has not changed its fiscal year during this period.

    The trading symbol of the companies has not been stopped for more than three months.

    The financial information they need is available. All listed companies have different fiscal years.

    Introduction

    The portfolio management process is a coherent set of steps that make the right portfolio to achieve investment goals and maintain it. It will be very useful and valuable to build a portfolio that is in accordance with the level of risk tolerance of the investor and his personal preferences, and in addition, it is a collection of assets that have favorable risk and return characteristics. Portfolio management is a multi-dimensional problem, and a multi-criteria decision-making approach provides methodological foundations for solving the multi-criteria nature of the problem. Using technical and fundamental analysis simultaneously, in addition to investors' preferences, can be very appropriate. In addition, due to the fact that investors have relative information about the market and have to face a high level of uncertainty, also the interaction between technical and fundamental criteria is not certain, the problem is complicated and uncertain and is not structured. Therefore, the use of artificial intelligence techniques is very beneficial. Among these techniques, the rule-based expert system is a very suitable framework for problem solving. Financial planning is what most people do. Before people start investing, each person should prepare a general financial plan. Such a plan should include making a decision regarding the transaction. In addition, the method of ownership, the life span of the asset and its profitability should also be considered. Finally, this plan should include the minimum amount of necessary savings. Investors seek to manage and enhance their wealth and assets by investing in an optimal mix of financial assets. The concept of optimal combination is important, because people's wealth, which is kept in the form of different assets, should be evaluated and managed as a unit. Wealth should be managed and evaluated in the form of a portfolio. A portfolio includes an investor's investment collection.

    2-1-2 Why do we invest?

    Investment is made to make money. Although everyone agrees with this statement, we need to be more precise about it. We invest to improve our current and future well-being.

  • Contents & References of Optimal portfolio selection using rule-based fuzzy expert system

    List:

    Table of Contents

    Page

     

    Title

     

     

     

    Chapter One: Outline of the Plan

    1-1 statement of the problem. 2

    1-2 research objectives. 2

    1-3 The importance of the research topic and the motivation for choosing it. 3

    1-4 research hypotheses. 3

    1-5 research model. 4

    1-7 research methods. 6

    1-8 scope of research. 6

    1-9 community and sample size. 6

    1-10 limitations and research problems. 7

    Chapter Two: Theoretical Studies

    Introduction. 9

    2-1 Theoretical research literature. 10

    2-1-1 Investment. 10

    2-1-2 Why do we invest?. 10

    2-1-3 The importance of investment study. 11

    2-1-4 investment process. 11

    2-1-5 The structure of the decision-making process. 12

    2-1-5-1 Securities analysis. 12

    2-1-5-2 portfolio management. 13

    2-1-6 Markowitz model. 14

    2-1-7 Criticisms on the modern portfolio theory. 16

    2-2 expert systems 17

    2-2-1 different definitions of expert system 20

    2-2-2 advantages of expert systems 20

    2-2-2-1 organizational advantages. 20

    2-2-2-2 User benefits. 23

    2-2-3 Limitations of expert systems 23

    2-2-4 Expert systems and artificial intelligence. 23 2-2-5 General method of expert systems analysis (initiative method) 26 2-2-6 General model of expert systems 26 2-2-7 Elements of an expert system 27 2-2-8 Applications of expert system 28 2-2-9 Use of rules to represent knowledge. 30

    2-2-10 fuzzy expert systems. 30

    2-2-11 review of the software used for expert system design 32

    2-2-12 Comparison of expert systems and decision support systems. 33

    2-2-13 difference between DSS and expert systems 34

    2-3 stock market analysis. 35

    2-3-1 Fundamental analysis. 35

    2-3-1-1 strengths of fundamental analysis. 38

    2-3-1-2 weaknesses of fundamental analysis. 39

    2-3-1-3 Some fundamental indicators used 39

    2-3-2 What is technical analysis?. 47

    2-3-2-1 Philosophy or logic. 49

    2-3-2-2 indicator (index) 50

    2-3-2-3 What does an indicator provide? 50

    2-3-2-4 Why do we use indicators?. 51

    2-3-2-5 tips about using technical indicators. 51

    2-3-2-6 leading indicators (guiding) 52

    2-3-2-7 regressive indicators (consolidating) 53

    2-3-2-8 leading indicators versus regressive. 53

    2-3-2-9 Oscillation of Momentum Views (measure of movement) 54

    2-3-2-10 Types of Oscillation Views 54

    2-3-2-11 Some important terms. 55

    2-3-2-12 terms used in the field of price. 59

    2-3-2-13 Advantages and disadvantages of technical analysis. 60

    2-3-2-14 Flexibility and adaptability of technical analysis. 61

    2-3-2-15 technical analysis and its efficiency in different time frames. 62

    2-3-2-16 economic forecasts. 62

    2-3-2-17 Some criticisms about technical analysis. 63

    2-3-2-18 How to coordinate technical and fundamental analysis? 67

    2-3-2-19 technical indicators used 68

    2-4 research background. 74

    2-4-1 Portfolio management. 74

    2-4-2 Financial analysis. 77

    2-4-3 Financial planning. 78

    2-4-4 Rating of bonds. 78

    2-4-5 Bankruptcy risk assessment. 79

    2-4-6 Investment. 79

    2-4-7 Credit granting. 80

    Chapter 3: Research identification method (methodology)

    Introduction. 82

    3-1 research method. 82

    3-2 Statistical population. 83

    3-3 sample volume and measurement method. 83

    3-4 Information collection tools. 83

    3-5 Data analysis method 84

    3-6 Hypotheses 85

    3-7 Variables 85

    3-7-1 Fundamental indicators used 85

    3-7-2 Technical indicators used 86

    3-9 Research model. 87

    Chapter Four: Analysis of Research Findings

    Introduction. 89

    4-1 The first step of removing unacceptable shares. 89

    4-2 stock evaluation stage. 90

    4-2-1 Fuzzification of inputs 98

    4-2-2 De-fuzzification. 101

    4-4 The final portfolio. 104

    4-5 Analysis of hypotheses104

    4-5 analysis of research hypotheses. 126

    4-5-1 The first hypothesis. 126

    4-5-2 The second hypothesis. 126

    4-5-3 The third hypothesis. 127

    Chapter Five: Conclusions and Suggestions

    Introduction. 138

    5-1 Conclusion. 138

    5-2 Research limitations and problems. 141

    3-5 suggestions from the results of research findings. 141

    5-4 Suggestions for future research. 142

    List of Persian sources. 144

    List of English sources. 145

    English abstract. 149

    Source:

    List of Persian sources

    O'Brien, James. A. Manian, Amir, Fatahi, Mehdi, Vasheq, Bahareh, Management Information Systems, Negah Danesh Publications, 1386

    Elahi, Shaaban, Rajabzadeh, Ali, Expert Systems for Intelligent Decision Making, Business Publishing Company, First Edition, 1382

    Bakhshiani, Abbas, Rai, Reza, Stock Valuation and Market Analysis, Industrial Management Organization, 1387

    Barzowi, Rajabali, Discussions in Fuzzy Set Theory (Collection of Articles), Zahedan: University of Sistan and Baluchistan, 1st edition, 1381

    Tehrani, Reza, Financial Management, Negah Danesh Publications, 6th edition, 1388

    Jones, Charles P., Tehrani, Reza, Nourbakhsh, Asgar, Investment Management, Negah Danesh, 8th edition, 1391

    Khaki, Gholamreza, research method with an approach to writing a thesis, Reflection Publications, 4th edition, 2017

    Rafiei Imam, Alineqi, selecting the best stocks by fundamental analysis method, Nass Publications, first edition, 2017

    Sarmad, Zohra, Bazargan Herandi, Abbas, Hejazi, Elaha, research methods in behavioral sciences, Aghaz Publications, 6th edition, 2011

    Shodhaei, Seyyed Mohammad Ali, Fundamental analysis in the capital market, Challenge publication, 4th edition, 1390

    Latfi, Ali, Darwish, Zahra, Technical indicators, Terme publishing house, 1385

    Murphy, John, Technical analysis in the capital market, Farahani Fard, Kamiar, Ghasemian Langroudi, Reza, Challenge publication, 2nd edition, 1384

    Mirez, Thomas, stock price prediction in Bruce by technical analysis method, Sami Far, Shadi, Shaban Ali, Mohammad Reza, Nass Cultural Institute, third edition, 1390

    List of English sources

    Achelis, Steven B., Technical Analysis from A to Z, McGraw-Hill, 2000

    Bahrammirzaei, Arash, A comparative survey of artificial intelligence applications in finance: Artificial neural networks, expert systems and hybrid intelligent systems, Neural Comput & Applic, 19, 2010

    Bao, Depei, Yang, Zehong, Intelligent Stock Trading System by Turning Point Confirming and Probabilistic Reasoning, Expert Systems with Applications, 2008, 34

    Darlington, Keith, The essence of expert systems, Prentice Hall, 2000

    Elmer, Peter J., Borowski, David M., An Expert System Approach to Financial Analysis: The Case of S&L Bankruptcy, Financial Management, Autumn 1988, Vol. 17, No. 3, pp. 66-76

    Fasanghari, Mehdi, Montazer, Gholam Ali, Design and Implementation of Fuzzy Expert System for Tehran Stock Exchange Portfolio Recommendation, Expert Systems with Applications, 2010, 37, pp. 6138-6147

    Giarratano, Joseph C, Riley, Expert systems: principles and programming, PWS-KENT Pub. Co., 1989, third edition

    Harmon P, King D, Artificial intelligence in business expert systems, Wiley, New York, 1985

    Kim, Bonn-Oh, Lee, Sang M., A Bond Rating Expert System for Industrial Companies, Expert Systems with Applications, 1995, Vol. 9, No. 1, pp. 63-70

    Lee, Jae B., Stohr, Edward A., Representing Knowledge for Portfolio Management Decision Making, Center for Digital Economy Research Stem School of Business, 1985

    Lee, Jae Kyu, Kim, Hyun Soo, Intelligent Stock Portfolio Management System, Expert Systems, April 1989, Vol. 6, No. 2

    Lee, Jae Kyu, Nam, Sang Zo, An Object-oriented Optimal Savings System: HYPER-SAVINGS, Intelligent Systems in Accounting, Finance and Management, 1997, Vol. 6, pp. 303-320

    Lee, K. H., Jo, G. S., Expert System for Predicting Stock Market Timing Using a Candlestick Chart, Expert Systems with Application, 199, 16

    Liu, N. K., Lee, K. K.

Optimal portfolio selection using rule-based fuzzy expert system