«Kaunas University of Technology Andrius Jurgutis DEVELOPMENT AND INVESTIGATION OF MULTI-AGENT EXPERT INVESTMENT DECISION SUPPORT SYSTEM Summary of ...»
Kaunas University of Technology
DEVELOPMENT AND INVESTIGATION
OF MULTI-AGENT EXPERT
INVESTMENT DECISION SUPPORT
Summary of Doctoral Dissertation
Technological Sciences, Informatics Engineering (07T)
Doctoral dissertation was prepared at Kaunas University of Technology, Faculty
of Electrical and Control Engineering, Department of Process Control in 2008Scientific Supervisor:
Prof. Dr. Habil. Rimvydas SIMUTIS (Kaunas University of Technology, Technological Sciences, Informatics Engineering – 07T)
Dissertation Defence Board of Informatics Engineering Science Field:
Chairman Prof. Dr. Robertas DAMAŠEVIČIUS (Kaunas University of Technology, Technological Sciences, Informatics Engineering – 07T).
Prof. Dr. Habil. Feliksas IVANAUSKAS (Vilnius University, Physical Sciences, Informatics – 09P), Assoc. Prof. Dr. Arnas KAČENIAUSKAS (Vilnius Gediminas Technical University, Technological Sciences, Informatics Engineering – 07T), Prof. Dr. Habil. Genadijus KULVIETIS (Vilnius Gediminas Technical University, Technological Sciences, Informatics Engineering – 07T), Assoc. Prof. Dr. Olga KURASOVA (Vilnius University, Technological Sciences, Informatics Engineering – 07T).
Prof. Dr. Adas GELŽINIS (Kaunas University of Technology, Technological Sciences, Informatics Engineering – 07T), Prof. Dr. Habil. Aleksandras Vytautas RUTKAUSKAS (Vilnius Gediminas Technical University, Social Sciences, Economics – 04 S).
The official defence of the Dissertation will be held at the open meeting of the Board of Informatics Engineering Science Field at 10 a. m. on 30 October, 2014 in the Dissertation Defence Hall of the Central Building of Kaunas University of Technology.
Address: K. Donelaičio str. 73, room no. 403, LT-44029 Kaunas, Lithuania.
Tel.: +370 37 30 00 42; fax +370 37 32 41 44; e-mail: email@example.com The summary of the doctoral dissertation is sent on 30 September 2014.
The Dissertation is available at the library of Kaunas University of Technology (K. Donelaičio str. 20, Kaunas, Lithuania).
Kauno Technologijos Universitetas Andrius Jurgutis
INVESTICINIŲ SPRENDIMŲ PARAMOS
SISTEMOS KŪRIMAS IR TYRIMAIDaktaro disertacijos santrauka Technologijos mokslai, informatikos inžinerija (07T) 2014, Kaunas Disertacija rengta 2008-2013 metais Kauno Technologijos Universiteto Elektros ir valdymo inžinerijos fakultete, Procesų katedroje.
Prof. habil. dr. Rimvydas SIMUTIS (Kauno Technologijos Universitetas, technologijos moksliai, informatikos inžinerija – 07T)
Disertacija ginama informatikos inžinerijos mokslo krypties taryboje:
Pirmininkas prof. dr. Robertas DAMAŠEVIČIUS (Kauno technologijos universitetas, technologijos mokslai, informatikos inžinerija – 07T).
prof. habil. dr. Feliksas IVANAUSKAS (Vilniaus Universitetas, fiziniai mokslai, informatika – 09P), doc. dr. Arnas KAČENIAUSKAS (Vilniaus Gedimino Technikos Universitetas, technologijos mokslai, informatikos inžinerija – 07T), prof. habil. dr. Genadijus KULVIETIS (Vilniaus Gedimino Technikos Universitetas, technologijos mokslai, informatikos inžinerija – 07T), doc. dr. Olga KURASOVA (Vilniaus Universitetas, technologijos mokslai, informatikos inžinerija – 07T).
prof. dr. Adas GELŽINIS (Kauno Technologijos Universitetas, technologijos mokslai, informatikos inžinerija – 07T), prof. habil. dr. Aleksandras Vytautas RUTKAUSKAS (Vilniaus Gedimino Technikos Universitetas, socialiniai mokslai, ekonomika – 04 S).
Disertacija bus ginama viešame Informatikos inžinerijos mokslo krypties tarybos posėdyje 2014 m. spalio mėn. 30 d. 10 val. Kauno Technologijos Universiteto centrinių rūmų disertacijų gynimo salėje.
Adresas: K. Donelaičio g. 73, 403 a., LT-44029 Kaunas, Lietuva.
Tel.: (8 37) 30 00 42; faksas (8 37) 32 41 44; el. paštas firstname.lastname@example.org Disertacijos santrauka išsiųsta 2014 m. rugsėjo mėn. 30 d.
Disertaciją galima peržiūrėti Kauno Technologijos Universiteto bibliotekoje (K.
Donelaičio g. 20, Kaunas).
INTRODUCTIONDefinition of the problem Rapid development of information-oriented society and its environment launches a lot of challenges in finance and investment management sphere. World financial markets have become easily accessible for investors from all over the world owing to modern information technologies. Easy access to world financial markets has encouraged the growing of the investor society, which in its turn exerts strong influence on world economic development. Modern investors are displeased only with the opportunity of easy access to financial markets, they need information and knowledge. Having easily reached financial markets investors directly face the difficulty of information processing and management, i.e.
financial markets are distinguished for rapidly changing information environment.
With plenty of such information at hand both individual and institutional investors experience difficulties covering such huge amount of information. Even having coped with plenty of information there is still a very challenging task left to obtain knowledge from it. Financial markets is a place where the obtaining of knowledge from information present at financial markets must take place in real time and quickly. Time is very important for investment success.
Raised issues are of practical nature, and their settlement must involve different scientific knowledge such as economics, financial management, mathematics, information technologies etc. In summary it can be said that modern investors need all-in-one solution (combining the management of both information and knowledge) which can simplify their orientation in financial markets and could be implemented through the medium of modern investment management or investment decision support information system. In such a way work problematic comes down to the creation of investment decision information system.
Topicality of the research Nowadays in order to invest, to trade and to speculate at financial markets successfully it is insufficiently only to be sound on fundamental, technical or other methods of securities assessment and portfolio management. Today, when the trade of securities has moved into the internet environment investors face new challenges.
First of all, financial markets have become global, i.e. on the one hand, it has become very easy for investors to make investments in any world region, but on the other hand, they face a need to be sound on the specific character of each corresponding region, adapt to it and learn to work there. Secondly, financial markets have become dynamic, i.e. they can quickly change directions, thus there is a need to follow information continuously, to analyse it, to draw immediate conclusions and make investment decisions on the ground of them. Consequently, investment management is rather complex and multipronged task. Naturally appears a need to create intellectual information systems able to help an investor to make investment decisions.
Classical companies which consult investors and administer their property work employing the teams of specialists of different spheres. Some of them search, collect and group information, others analyse it and draw conclusions. The third forecast future scenarios, the forth summarize and draw final conclusions. It is thought that such work pattern in the century of the internet and other advanced information technologies does not simply correspond not only to the needs and expectations of modern investors but also to the interests of the owners of these companies. Subject to the existence of intense competition among the consulting and property management companies there is a wish to optimize corporate expenses. One of the possible solutions suggests computerizing the part of work executed by specialists, i.e. to create an intellectual information system able to consult investors.
Such a system would be useful both for an investor and a company rendering such kind of services. Firstly, it would let to increase the profitability of consulting and investment management companies. Secondly, it would allow rendering cheaper services to investors and achieve smaller expenses for different investment funds management. Thirdly, computer-assisted system would quicken decision making what is very important at the dynamic market. Investment practice shows that greater return is achieved by those who get, find, correctly evaluate information and come to a decision the first. Plenty of information sources and their formats aggravates and prolongs information perception and its processing time. Thus, modern investment management information system must be able to interpret information, find it, compare and update it. Fourthly, computerassisted consulting system could ensure greater investor confidentiality and execute preventive market manipulation work. Fifthly, intellectual consulting system would manage to adapt to the changing investor’s goals. In such a way a client would always be hedged round with attention and fair offers. At the same time it would be a new leap and standard in client servicing quality as well as newgeneration method of investment ideas and investment products sale. Sixthly, it would be possible to transfer the best experience of various specialists into an intellectual consulting system thus avoiding possible human failings. Seventhly, a consulting information system possessing all the above mentioned qualities would increase the number of correct decisions in comparison to classical consulting companies. Eighthly, in case of the success of the implementation of such intellectual investment consultant there is a strong probability to attract more people interested in individual investment or wishing to try it. Moreover, it would allow an investor to know the investment world faster.
Investment decision making, investment portfolio management is a complex and multipronged task. Complexity appears when one looks for information scattered on the internet or elsewhere, sorts out the methods of securities evaluation and finally draws conclusions. Complexity is that in order to make a decision it is necessary to coordinate the set of coherently and in parallel related processes such as information gathering, carrying out of evaluation, market surveillance etc., and to establish interrelations. Thus, providing of investment recommendation to an investor comes down to the solution of some tasks and to the control and management of separate tasks. It is assumed that it is reasonable to create such a system using software agents or rather to make multi-agent systems of them for task solution.
The object of the scientific research The object of the scientific research is a multi-agent autonomous expert investment decision support system that forms real-time investment decisions at financial markets and helps to form and manage investment portfolio.
The objective of the work Work objective is the easement of investment problem solutions for different risk tolerance investors, the facilitation of the process of investment decision making and real time investment decision making applying multi-agent expert investment decision support system.
The tasks of the work In order to achieve the objective of the work the following tasks are put
forward and solved:
1. To investigate and choose the methods of the valuation (assessment) and forecasting methods of financial instruments for multi-agent expert investment decision support system.
2. To investigate and choose or develop the methods of investment portfolio formation and management for multi-agent expert investment decision support system.
3. To identify and choose the suitable software agent’s platform to implement multi-agent expert investment decision support system.
4. To create the following methods:
a) The method of the evaluation of investor risk tolerance, the method of the portfolio property allocation and the method of the providing of investment recommendation for securities;
b) To create the methodology of an investment decision support system implemented by software agents.
5. On the ground of the experiment outcome to identify and evaluate the opportunities and the main characteristics of multi-agent expert investment decision support system.
Methodology of the research The objective of the scientific work and the tasks put forward and solved for its achievement include various scientific spheres beginning with social sciences and finishing with exact sciences. So, the work uses a number of research methods. The analysis of literature related to the scientific work theme included the analysis, synthesis and generalization of scientific sources. The analysis of the questions related to the evaluation of securities, consistent pattern of financial markets and portfolio theory models included the methods of logical, comparative and system-oriented analyses. In order to transfer expert knowledge into the system the means and methods of computational intelligence creation were used.
For the specification, formation and implementation of a multi-agent investment decision support system generalization, synthesis and concretization methods were used. For the evaluation of system provided results and the possibilities of system itself the methods of mathematical-statistical analysis, simulation modelling, comparative and dimensional analysis were applied.
Scientific and engineering novelty of the work The prototype of the multi-agent expert investment decision support system presented in the work was implemented offering new scientific and engineering ideas. The system is original and has the following distinctions in comparison to
other similar systems:
1. Implemented in the system an original cascading expert fuzzy logic system is designed to evaluate investor risk tolerance.