Pages: 8

This introduction presents an overview of the key concepts discussed in the subsequent chapters of this book. The book introduces the way to analyze and design a risk analytics system that integrates the components of what risk analytics does in separated settings using multiple approaches to deal with a diverse type of data and problems to solve. It provides the fundamentals of risk management, mathematics, traditional models in finance, and the basis of analysis and design of information systems. The book presents how risk management and analytics are acting in organizations, and deals with several aspects in risk analytics application to solve problems. It explores tools for risk management: analytics and non-analytics based. The book explains the basic concepts of techniques in machine and statistical learning in risk analytics. It indicates how risk analytics provide support to monitoring risks, dealing with bias, implementation barriers, and assumptions in the risk analytics process.

Pages: 41

This chapter describes about the concepts that are required to build a risk analytics system. The nature of risk analytics is that techniques and processes are applicable to any type of organization. In its implementation, risk analytics faces important challenges, such as data management, dealing with assumptions, the modeling process, the standards and metrics building process, the evaluation process, and communication among stakeholders. Through learning, the accumulation of knowledge is a tool to manage uncertainty when risk is considered as a variation of results and knowledge as a means to reduce uncertainty. Digging deeper in risk analytics and the enterprise risk management approach there is an orientation to connect operations management and the decision-making process with risk theory that are related to randomness and uncertainty. From the probability point of view, risk analytics will apply different approaches depending on the variable type.

Pages: 47

This chapter provides a review of the steps in the risk management process, the risk modeling process, and the way to create a risk analytics road map. The risk modeling process needs to mobilize information flows to produce risk knowledge to consolidate, integrate, and organize the risk knowledge across organization and risk-related problems to solve. This can happen in the following way: risk knowledge creation where risk knowledge is represented by risk assessment; risk knowledge storage and retrieval through the data support for external and internal users; risk knowledge transfer using the experience of many people in known cases; and risk knowledge application to discover business processes opportunities. The exploratory data analysis starts with graphic representation (data visualization) and needs to move to summary statistics where the shape of variable distributions is observed and the main metrics moments of the distribution are analyzed in an aggregate of separated way risk units.

Pages: 49

This chapter introduces general concepts of decision making and how to measure through risk indicators using several techniques. It frames the general metrics development about basic methods/tools of decision making and describes about general risk metrics used and its benchmark. The chapter compares results of the metrics, means, and variances in different groups, products, regions, and deals with multiple decisions and how to forecast possible results. It presents the examples supplement the general review of how the analytics process and the risk management process are connected. In general, in risk analytics the use of key performance indicators and key risk indicators has several facets: first, the identification of groups, classes, and categories in the factors used to define the strategy matrices. A second aspect is related to the definition of factors to evaluate and compare the way to obtain the scores.

Pages: 35

This chapter comprises a review of the risk measurement process as a component of a risk analytics system creation supplementing. A risk measurement system is integrated to the planning process in all levels of the organization, and it includes the metrics definition, data sources selection, validation of models, stress testing, and several testing evaluation interactions depending on the risk types. The metrics development such as probabilities of default, default rates analysis, or metrics scales definition are part of the risk exposure understanding and the guide to risk analytics for risk monitoring and control of the risk management processes. An illustration about the real risk measurement process is given through the banks approach. Risk appetite is part of an effective risk management to protect the organization from unacceptable losses or undesirable outcomes with respect to earnings volatility, capital adequacy or liquidity, reputation risk, or other risks while supporting and enabling the overall business strategy.

Pages: 62

This chapter provides a presentation of the risk analytics process to be implemented in a risk analytics system design following two approaches. First, qualitative models and improvement of knowledge to deal with uncertainty and manage risk and second, to connect the approaches of dealing with risk through financial concepts of risk management based on return-interest theory, portfolios, and products for risk management (derivatives). The analysis of return and risk management indicates some of the most common tools on the market to cover investments or assets against risks of different kinds. Risk analytics has a lot of room to grow and to connect the risk knowledge creation and application to create value and to improve risk management processes and systems. There is a description of managing risk knowledge as an instrument of strategic development in organizations and how return analysis in the traditional way has been used for dealing with risk.

Pages: 19

This chapter describes about the type of data that is required to use in risk analytics according to the variety of data. It shows that risk analytics data can be structured, the traditional data coming from data warehouses/relational databases; data from other types of human actions, speech, pictures, videos, text, with batches and continues actualization such streaming. The chapter explains about what is expected and the directions of using technology for risk analytics, product development, and support organization and society in areas related to the risk management process. Risk analytics requires a systematic and methodic data gathering and organization process. There are many ways to describe the value that cloud computing brings to the organization of risk management. The risk analytics system connects multiple points that are in different levels of development and with possibilities to increase capabilities for organizations.

Pages: 93

This chapter provides an introduction to a variety of analytics techniques used in risk analytics that are essential for the risk analytics system analysis and design. It exposes the statistical and machine learning techniques that are used in combination to the previous models that are leading to solve problems in pricing, valuation, transactions, etc. The chapter presents the use of the techniques to support knowledge discovery through description, classification, optimization, simulation, etc. These techniques supplement the direct creation of metrics by products or stages in a risk management process. The chapter also presents the techniques that are part of the arsenal of statistical and machine learning tools and are selected to guide the reader to the relationship type of problem to solve and possible technique to use. However, there is not an extensive review of many algorithms because the number of packages/algorithms is growing very fast and expanding immensely, but the basic group of models is exposed.

Pages: 17

The risk analytics process is intimately related to monitoring risk or the construction of a complete intelligent system to perform the risk management processes. This chapter focuses on implementation and adoption of the risk analytics processes. It reviews the concepts in related to the understanding of means to embed risk management to strategic planning and the review of possible barriers that are present to maintain a risk analytics system (RAS) properly operating. The chapter describes components of the RAS needs the context where people and machine interact. The business environment guides the definition of the RAS components for organizing and connecting key performance indicators and key risk indicators, and data required, as it was introduced models to analyze data, and means to transfer and keep the knowledge generated by the risk analytics process. Each area of metrics definition, RAS design, and risk analytics work is operating according to control subsystems.

Pages: 73

This chapter provides a presentation of examples mainly in financial markets/institutions related to concepts regarding the structure of the risk analytics system (RAS). A RAS is part of a general risk management information system that is connected to operations credit, investment, marketing, operation, etc. and general operations of the organizations. In the RAS design, the knowledge management processes are the core of the system. The risk analytics example illustrates how to deal with data and Monte Carlo Simulation as an important component that the RAS should have. The creation of actions in risk analytics is translated to the creation of information systems that contain a data analytics engine to describe, predict, and prescribe the operations in businesses under uncertainty observation. The analysis of risk has been in many organizations out of the target of the methodologies that are associated with productivity, in particular to add results to the business value.