engleski

Capital Requirements and Measuring Market Risk in EU New Member States and Croatia in light of Basel Committee Guidelines

The dissertation is composed of eight parts. The first part of the dissertation is the introduction to the doctoral dissertation. In this part the problem and the subject of the dissertation, purpose and the goals of the research, along with the scientific hypotheses, methods used, scientific contribution and structure of the dissertation are presented. The second part of the dissertation serves as an introduction to the nature and theory of financial risk. The basic notions in risk management are presented as well as the explanation and description of the basic forms of risks encountered in banking business. The third part present the development of current Basel rules for measuring and managing market risk in banks. The two available approaches of calculating provisions for market risk are presented. Main advantages and disadvantages of both the standardized and internal approach are analysed and discussed. This part of the dissertation concludes with an overview of risk measurement and management importance, characteristics and practices in banking sector of EU new member states and Croatia. The fourth part of the dissertation deals with the Value at Risk as a method of calculating capital charge for market risk. The definition, historical development and the rationale behind Value at Risk is presented. Possible opportunities for using Value at Risk method beyond market risk measurement are presented and discussed. The fourth part concludes with the presentation of the main advantages and disadvantages of Value at Risk as a method of measuring market risk. The fifth part represents the theoretical basis for the understanding and development of Value at Risk models. This part of the dissertation begins with the explanation of time series models used in financial engineering and risk management. Special attention is given to generalized autoregressive heteroskedasticity (GARCH) models. The fifth part continues with the theoretical explanation and rationale of parametric and nonparametric models to calculating VaR. In the last part of the dissertation a family of semi parametric VaR models is presented. In this part of the dissertation the author develops a new semi parametric VaR model, called Hybrid Historical simulation. The sixth part of the dissertation deals with backtesting of the VaR forecasts and introduces the best-known and newest methods of backtesting and evaluating VaR estimates. Advantages and disadvantages of each backtesting procedure are discussed. The seventh part of the dissertation is the empirical analysis of selected VaR models presented and discussed in the dissertation. The summary of empirical research into VaR is presented. Characteristics of analysed stock indexes from EU new member states and Croatia are presented and obtained results are discussed. In this part of the dissertation the data, methodology, as well as VaR and volatility forecasting models used in the testing of stock indexes are explained in detail. The seventh part continues with the actual backtesting across different criteria of the selected VaR models as well as the new VaR model developed by the author in the Chapter 5 of the dissertation. This part concludes with the presentation and discussion of the obtained results. The last part of the dissertation sums up the main finding and results of the research.

Capital requirements; market risk; EU New Member States; Croatia; Basel 2; Value at Risk; Hybrid historical simulation

nije evidentirano