Thesis Details
FORMAL MODEL OF DECISION MAKING PROCESS FOR HIGH-FREQUENCY DATA PROCESSING
This thesis deals with the issue of the processing of high-frequency time series. It primarily focuses on the design of algorithms and methods for support of predicting these data. The result of this work is a model supporting the decision-making process implemented into a complex platform. The model designs the method of formalization of business rules which describes the decision-making process. The designed model must meet the conditions of the robustness, scalability, real-time processing and econometrics requirements. The thesis summarizes the current knowledge and methodologies for the processing of high-frequency fnancial data which can be found on the stock exchange.The first part of the work describes the basic principles and approaches currently used in the processing of high-frequency data. The next part deals with the description of an appropriate complex event platform and is subsequently devoted to prediction and data processing itself, using the chosen platform. Emphasis is on selecting and editing a set of rules that controls the decision-making process. The newly designed method describes the set of rules by using matrix grammar. This grammar belongs to the grammars with regulated rewriting and thus it may control the data processing by the defning of the matrices.
High-frequency data, CEP, time series, business rules, decision-making process, real-time processing, formalization, Esper.
@phdthesis{FITPT645, author = "Eva Z\'{a}me\v{c}n\'{i}kov\'{a}", type = "Ph.D. thesis", title = "FORMAL MODEL OF DECISION MAKING PROCESS FOR HIGH-FREQUENCY DATA PROCESSING", school = "Brno University of Technology, Faculty of Information Technology", year = 2017, location = "Brno, CZ", language = "english", url = "https://www.fit.vut.cz/study/phd-thesis/645/" }