Enterprise Risk Management Solutions - Delivering a Solid Foundation

Enterprise Risk Management Solutions Delivering a Solid Foundation A SAS White Paper Table of Contents Executive Summary .......................................................................................................1 Introduction ....................................................................................................................1 Sourcing Data for Analysis ...........................................................................................2 Standardization of Data ................................................................................................2 Cleansing and Quality Checks .....................................................................................3 Data Quality Tools.....................................................................................................4 Data Quality within Six Sigma Projects.....................................................................5 Data Deficiency .........................................................................................................5 Creation and Documentation of Metadata....................................................................6 Technical Metadata Layer.........................................................................................6 Business Metadata Layer .........................................................................................7 Metadata Management .............................................................................................8 Transforming Data into Information.............................................................................9 Information Marts ......................................................................................................9 Exploiting the Results ...............................................................................................9 Loading the Warehouse .............................................................................................10 Warehouse Administration .........................................................................................10 Conclusion ....................................................................................................................11 Enterprise Risk Management Solutions Executive Summary The greatest challenge facing an enterprise risk management (ERM) implementation is the movement of large amounts of data from a myriad of sources to a well-documented, consistent, friendly format. Disparate data sources form an often overwhelming hurdle for anyone involved in the implementation of an ERM solution. Challenges include everything from the sheer number of sources to ensuring the integrity of the staged data. The process of systematically organizing information to support an enterprise risk initiative is fundamental to ensuring good risk management practices. Warehousing methods found within the SAS® Risk Dimensions Solution can help technologists deal with the many data extraction, transformation and loading obstacles risk-related projects must overcome. SAS Risk Dimensions provides a high- performance and scalable architecture designed to deliver a substantial return on investment (ROI). SAS technology is already used by 98 of the top 100 companies in the Fortune 500 to create business intelligence from raw data. SAS Risk Dimensions can help turn a variety of data into useful information. Introduction Similar to constructing buildings, ERM implementations require a solid foundation. But unlike construction, a properly designed warehouse must constantly change and evolve to keep pace with the business climate. A warehouse should allow the information that is unique to your business to be surfaced quickly. Applying risk analyses to the source data and combining a robust suite of risk analysis techniques with a solid warehousing process lets users identify trends and take the best action before the firm is affected. This proactive approach to enterprise risk management can deliver substantial ROI to the firm. To realize that ROI, the foundation must be capable of supporting a wide variety of user requirements. Some of these requirements include: · Seamless access to information and data stores across the enterprise. · Executive dashboards. · Executive scorecards. · Ad hoc risk analysis environments. · Daily risk reporting requirements. · Robust risk analytics. · Regulatory analysis. · Historical analysis. The process of planning, designing and implementing a warehouse that is capable of supporting an ERM platform is critical to success. Building a warehouse to support these 1 Enterprise Risk Management Solutions business requirements typically calls for the coordinated efforts of the business and technology groups. Similar to a building, the foundation must support the requirements of the end users. The transformations and business rules needed to turn unorganized data into information is unique for each line of business. Coordinating this effort between business and IT is critical to defining the high-level goals of the project correctly. This list is not the end result nor will it provide a static list of requirements. These lines of communication provide a mechanism for the development of a platform that is capable of supporting the business needs as they evolve. The ultimate design of the foundation must survive in a sometimes chaotic infrastructure. SAS provides a methodology and technology for users to see results quickly from an integrated risk warehouse and reporting infrastructure. SAS’ data warehousing methodology includes seamless data access, centralized metadata creation, transformation of data into risk related knowledge and IT administration that substantially increases the ROI of a risk-related project. Sourcing Data for Analysis Three major components are required to obtain data from source systems and organize it successfully into a risk data warehouse: · Standardization of data. · Cleansing and quality checks. · Creation and documentation of metadata. The SAS Risk Dimensions Solution integrates information from a multitude of sources — providing native access to more than 65 databases, messaging processes, third-party sources, the Web and more — enabling you to gain a holistic view of your data. By accessing data as a native client, SAS can implement a wider range of technology faster than traditional methods. Standardization of Data The majority of data required for an ERM solution resides in different transactional systems scattered throughout your organization. These systems span the technology spectrum from older legacy systems to complex Excel spreadsheets residing on users’ desktops. The data in these systems tends to be organized in a highly normalized format. In the simplest manner, a normalized format means there are multiple tables representing a contract. The example shown in Figure 1 is an illustration of a normalized table. This format is ideal for fast data collection. However, it is difficult for users to restructure the data to perform additional analyses. For example, if a user needs to calculate an exposure to a particular company, the data shown in Figure 1 is inadequate. The user would need to perform data manipulations to combine tables across lines of business. In addition, many of the reporting variables require additional analyses or computation. An ERM platform requires data that is formatted for easy reporting as well as for performing complex analysis. 2 Enterprise Risk Management Solutions Figure 1 – A normalized data table showing a list of FX Forward contracts for a particular transactional system. Cleansing and Quality Checks In an age of information, the effectiveness and accuracy of decision making is highly dependent upon the quality of information available. Therefore, another challenge for any warehousing project is the ability to deliver quality data. Figure 2 demonstrates a simple example of poor data quality. This table shows a counterparty name with multiple abbreviations and typos. Creating a single Counterparty ID from this table is difficult without first cleaning the variations. Large amounts of missing data or incorrectly entered data present more difficult challenges. In these cases, business users must help IT identify and articulate the business rules used to cleanse the data.