Centre for Modelling and Analysis of Big Data in Finance and Economics

Innopolis University
Scroll down

Russian Science Foundation Project No. 16-18-10432

Modern methods of robust inference in finance and economics, with applications to the study of crises and their propagation in financial and economic markets World-class research in statistical analysis and modelling of financial and economic processes, with applications to the study of crises and their propagation in financial and economic markets.

Project Page Summary (RU)

The study of financial and economic phenomena and markets and the dynamics of their key variables and indicators often requires statistical analysis of large databases of dependent and heterogeneous financial and economic observations such as high-frequency time series of financial returns and foreign exchange rates. Dependence, heterogeneity and outliers in analyzed data naturally appear due to volatility clustering in the dynamics of financial and economic indicators considered, the effects of crises and contagion in financial and economics markets. The problems of dependence, heterogeneity, outliers and large fluctuations in financial and economic variables and indicators considered significantly complicate their statistical analysis. Statistical inference methods that are often used in practice, including regression analysis and other widely applied approaches to estimation of parameters of financial and economic models, have to be modified under deviations from standard assumptions on independence and homogeneity in data that are very rarely satisfied in financial and economic applications. Unfortunately, many approaches to the analysis of dependent and heterogeneous data have poor statistical properties in finite samples that are typically observed in practice. Therefore, it is of significant interest to consider and focus on development and applications of new robust inference methods with better statistical properties. Further, important practical problems consist in incorporation of new and improved approaches to robust inference into widely used statistical software, development of new statistical software packages and toolboxes on their basis and the use of modern computer technologies in the analysis of properties of robust inference and their applications in the study of financial and economic phenomena and markets. In a similar way, a problem of considerable interest is given by applications of modern approaches to mathematical and computer modeling of processes and phenomena that give rise to problems of dependence, heterogeneity and outliers in financial and economic data. Undoubtedly, some of the most examples of such processes and phenomena are given by crises and their propagation on financial and economic markets and problems of financial contagion. The existing methods of the analysis of financial and economic markets often consider the problems of modeling crises and market interdependence separately from each other. In particular, powerful machinery for modeling and analysis of effects of crises on financial and economic indicators and markets is given by theory of heavy-tailed distributions. Together with development of its applications in finance and economics, the literature in these fields and its use in practice has focused on methods of modeling of interdependence of markets and properties of financial contagion based on theory of copulas, that is, functions that characterize all dependence properties of financial and economic variables considered. A further problem of significant theoretical and practical interest consists in development of methods of mathematical and computer modeling and statistical analysis of financial and economic markets that combine the above approaches. In addition to the above, an important problem in the modern literature in finance, economics, risk management and insurance is given by extensions of results for independent risks and random variables affected by crises to the case of dependence. The analysis. of the above problems can be based on computer and statistical modeling of wide classes of copula structures of interdependence of financial and economic variables with heavy-tailed distributions, including classes of copula models that generate tail dependence and, thus, financial contagion. An important problem of computer and statistical modeling and analysis on the basis of copula dependence models for financial and economic variables with heavy-tailed distributions is given by the analysis of robustness of properties of key models in finance, economics, risk management and insurance, including the study of (sub-)optimality of diversification under dependence and the effects of crises. The main goal of the present project is to make a significant advancement in the analysis of the above theoretical and applied research problems. This includes the development and applications of new and improved approaches to robust statistical analysis of big databases of dependent and heterogeneous financial and economic observations. It further includes the development and applications of modern approaches to computer and mathematical modeling of the effects of crises and interdependence of financial and economic markets simultaneously. One of the key aspects of the research on the project is given by the wide use of modern computer technologies and high-performance (parallel and distributed) computing systems, including that in important applications of the new approaches to robust statistical analysis and mathematical modeling of financial and economic markets as well as in the numerical analysis of properties of new methods and their comparisons with the existing inference and modeling techniques.

Problems of Key Importance and Interest

  • 1 Modern approaches to statistical and econometric analysis and modelling of financial markets, the dynamics of key financial indicators, their large fluctuations and crises.
  • 2 Development and applications of modern approaches to computer, mathematical and statistical modeling and analysis of the dynamics of key financial variables and indicators, crises and their propagation, financial contagion and interdependence in financial and economic markets.
  • 3 Development and applications of new and improved approaches to robust statistical analysis of big financial databases under the problems of non-linear dependence, volatility clustering, heterogeneity, large fluctuations, crises effects and financial contagion.
  • 4 Development and applications of modern mathematical and statistical models based on heavy-tailed distributions and copula structures in modeling and analysis of the dynamics of financial markets affected by crises and financial crises.
  • 5 Application of modern computer technologies, software and robust statistical inference methods for big data in the analysis of the dynamics of key variables and indicators in financial and economic markets in Russia, post-Soviet countries and transition and emerging economies
  • 6 Analysis of the effects of the on-going global financial and economic crisis and interdependence of markets in consideration on their development and the dynamics of their key variables and indicators.
  • 7 Špplication of modern computer technologies and software in modelling and robust statistical analysis of big databases on financial and economic markets affected by the crises, their propagation and financial contagion.
  • 8 Development and computer and statistical analysis of a large-scale databank on financial and economic crises that affected the Russian and post-Soviet economies. Computer and statistical modeling and analysis of the effects of financial and economic crises on the Russian economy and post-Soviet and emerging markets.
  • 9 Adaptation of modern methods for computer and mathematical modeling and robust statistical inference in financial and economic markets for efficient analysis using high-performance computing systems, and their software.

The main goal of the present project is to make a significant advancement in the analysis of the above theoretical and applied research problems. This includes the development and applications of new and improved approaches to robust statistical analysis of big databases of dependent and heterogeneous financial and economic observations. It further includes the development and applications of modern approaches to computer and mathematical modeling of the effects of crises and interdependence of financial and economic markets simultaneously. One of the key aspects of the research on the project is given by the wide use of modern computer technologies and high-performance (parallel and distributed) computing systems, including that in important applications of the new approaches to robust statistical analysis and mathematical modeling of financial and economic markets as well as in the numerical analysis of properties of new methods and their comparisons with the existing inference and modeling techniques.