Quantitative risk management and data analytics with applications to finance

Abstract

Risk Management has important implications for many organizations, and quantifying those risks is essential. The financial impact of the extreme events which lead to these risks is huge, a part of which is discussed in this thesis. We have various risk measures and properties associated with them. We studied risk measures in the context of market risk associated with equities and stylized facts of financial time series. We focused on the mathematical aspects of some of these stylized facts, which have implications for financial risk. These mathematical discussions have huge implications when working with situations where we have such nasty data that arise in many cases and thus implications of data analysis of these kinds of problems. We also examined the impacts of these new statistics on quantifying risk and capturing other aspects of financial time series. We also examined the dependence of these risk events and does it make sense to predict these extreme events based on past data and quantify their effects. We also discussed the economic and data analytic implications of the work. We also had a deeper look at how to make sense of the use of machine learning, some limitations of it, and how it affects our analysis and looked at the behavioral finance area from a new perspective. These insights may help open new horizons for further research in Machine learning, quantitative finance, behavioral finance, and other regions.