Advanced Methods of Analyses and Simulation

Code:PMM (FP RpmamP)
IT-MSC-2MMI-Compulsory-Elective - group N
Completion:classified accreditation
Type of
Hour/semLecturesSem. ExercisesLab. exercisesComp. exercisesOther
Guarantee:Dostál Petr, prof. Ing., CSc., II
Lecturer:Dostál Petr, prof. Ing., CSc., II
Faculty:Faculty of Business and Management BUT
Learning objectives:
  The aim of the course is to get acquainted with some advanced and non-standard methods of analysis and simulation techniques in economy and finance by the method of explanation of these theories, to become familiar with these theories and their use.
  The content of the subject is to make students familiar with the methods of analyses and simulation techniques (fuzzy logic, artificial neural networks, and genetic algorithms) by the way of explanation of the principles of these theories and their resulting applications in managerial practice.
Knowledge and skills required for the course:
  The knowledge in the area of math (linear algebra, arrays, analyses of functions, operations with matrixes) statistics (analysis of time series, regression analyses, the use of statistical methods in economy), operational analysis (linear programming), financial analyses and planning (the analyses of profits and costs, cash flow, value and bankruptcy model).
Learning outcomes and competences:
  The obtained knowledge and skills of the subject will enable the graduates the top and modern access in the processes of analyses and simulation in the national economy and private sector, organizations, firms, companies, banks, etc., especially in managerial, but also in economical and financial sphere.
Syllabus of lectures:
 1. Fuzzy logic (FL): To be familiar with the basic notions and fuzzy logic rules, creation of models. The presentation of cases of application of fuzzy logic in decision making processes e.g. managerial and investment decision making, prediction, etc.
2. Artificial neural networks (ANN): To be familiar with the basic notions in the area of artificial neural networks, presentation of the notation perceptron, multilayer neural network and their parameters. The applications cover investment decision making, estimations of the price of products, real properties, evaluation of value of client, etc.
3. Genetic algorithms (GA): To be familiar with the principles of genetics, the analogy between nature and math description that enables the solution of decision making of problems. The use in the area of optimization of wide spectrum of problems is mentioned - the optimization of investment strategy, production control, cutting plans, curve fitting, the solution of traveling salesman, cluster analyses, etc.
4. Chaos theory: The theory deals with the possibilities of better description of economic phenomena than the classical methods do. The notion chaos, order and fractal are clarified, the use of this theory to determinate the level of chaos of measured and watched system is mentioned
5. Data mining: The notion data mining, the definition of aims, the selection of methods of simulation, sources and preparation of data, creation of models, their verification, evaluation, implementation and maintenance are mentioned there. The presentation of the cases of the use for strategy of cooperation with customer, direct mailing, etc
6. Simulation: The presentation of the notion system and its identification and simulation. The description of the use of FL, ANN and GA during the process of simulation of decision making processes in enterprise sphere.
Progress assessment:
  The participation in lectures is not checked. The participation in trainings is compulsory and is systematically checked. The students are supposed to excuse for their absence. The teacher judges the reason of excuse. The way of substitution of a missed training will be set by the teacher individually.
Exam prerequisites:
  The credit will be granted in case of an active participation in trainings and handing in the final assignment, in case of need the written test. The work will range approximately from 8 to 12 pages concentrating on individual problem from practice leading to solution with the help of theory of fuzzy logic, artificial neural network or genetic algorithms.
The classified credit will be classified according ECTS. The way of implementation is in the form of test with in the range 0-20 points. A-20-19;B18-17;C16-15;D14-13;E12-;F10-0.