March 23, 2017 - Paul Lashmet | Industry Solutions

Visual Analytics for Fundamental Review of the Trading Book

The revised standards for minimum capital requirements, as laid out by the ‘Fundamental Review of the Trading Book,’ exemplifies the heightened regulatory expectations for the granular validation of risk models. This post describes visual analytics capabilities that will enable organizations to make the most of a data-driven regulatory program and establish a robust and defensible process across the enterprise.

What’s in a Name? Fundamental Review of Organization

Fundamental Review of the Trading Book (FRTB) is an initiative undertaken by the Basel Committee for Banking Supervision (BCBS) “to improve the overall design and coherence of the capital standard for market risk.” The result of this effort is the revised “Standards for Minimum Capital Requirements for Market Risk,” published in January 2016, replacing the 2006 rules. Below are key changes to the standards. A comment is added to each, emphasizing how they relate to an organization’s framework.

  1. The boundary between the trading book and the banking book is redefined in order to discourage the mechanization of avoiding capital costs through book to book transfers (regulatory arbitrage) – Evaluation of business models and governance processes.
  2. The Standard Approach (SA) to capital charge calculations is more reliant on risk sensitivities as inputs – Collaboration between front desk, risk, and finance functions.
  3. The approval process for a bank’s own set of risk models and capital calculations, called the Internal Model Approach (IMA), is more granular. Each desk is subject to supervisory approval and all capital charge models must be based on actual transactions with verifiable prices – Reorganization of the desk structure may be in order.

“Fundamental” is a good descriptor because, in order for an institution to comply with the new standards, there needs to be a fundamental review of how the organization operates.

[Related: “Review of Trading Book Remains Fundamental for Banks”]

Empower Subject Matter Experts with Enterprise Visibility

From a technology viewpoint, an organization must be able to hold and process large volumes of data from diverse sources and ensure its quality with regard to both accuracy and usefulness. The underlying calculations need to be hyper-efficient and adaptable because model performance will be a critical element of business decisions. AI and machine learning routines could be employed, dynamically generating even more data.

Organizationally, the new capital requirement for market risk standards requires enterprise visibility so that subject matter experts (“SMEs”) can work cross-functionally. SMEs must be empowered to assess the impact of capital charges and identify the drivers. Drivers of capital charges could range from trading strategy to incomplete data.

Visual analytics capabilities that facilitate cross-functional analysis are:

  • Showing real-time and historical data analysis in one application;
  • Analyzing all types of data in place, without moving it;
  • Allowing business users to do this in a self-service fashion.

Accelerate Actionable Insight to Help Decrease Capital Charges

An ISDA Quantitative Impact Study suggests that a large contributor to the expected increase in capital charges are Non-Modellable Risk Factors (NMRF), of about 30%. A risk factor is classified as “modellable” or “non-modellable.” A modellable risk factor requires a sufficient amount of transaction data, with “real” prices. The January 2016 standards define real prices as follows: (1) It is a price at which the institution has conducted a transaction; (2) it is a verifiable price for an actual transaction between other arms-length parties; or (3) the price is obtained from a committed quote. A modellable risk factor must be proven through extensive back testing. (This is related to Profit and loss (P&L) attribution and Backtesting. See BCBS explanatory note, p.5.)

The above illustrates the heightened expectations of regulators with regard to granularity of data quality and model validation. Depth of knowledge and comprehensive insight is required to mitigate the drivers of capital charges. Visual analytics capabilities that will accelerate actionable insight are:

  • Integrating machine learning for accelerated data quality evaluation and modeling;
  • Visually blending data that enables cross reference and correlation across various data sources;
  • Building real-time data apps for critical metrics and visuals that simplify time-series analysis.

Enable Efficient and Secure Cross-Functional Collaboration

Above, I described how subject matter experts (trading, risk, finance, compliance, etc.) need to be empowered with enterprise visibility to fully identify the drivers of capital charges. I also described the granularity of data needed to comprehensively assess impact. Here is an example:

  1. The Risk/Finance groups run a series of models;
  2. Risk/Finance advises the front office of how trading strategies affect capital charges;
  3. Traders, with this actionable insight, adjust as needed;
  4. All teams review and repeat until the right outcomes are achieved;
  5. Compliance and management monitor the process to ensure good governance.

However, think about this across multiple desks and regions and the infinite types of adjustments and alignments that can affect the bottom line. It is a dynamic, iterative process in which results are evaluated and assumptions are tested. Enterprise visibility and actionable insight are required to ensure a robust and defensible process.

As an example, the image below visualizes how multiple trading desks perform against a series of risk models. The process would be as follows:

  1. Test outcomes by adding/removing asset classes, individual positions, or entire desks;
  2. Visualize how the tests affect performance;
  3. Review the calculated results;
  4. Drill down to transaction details, ensuring validation of models;
  5. Build up analysis from one contributor to the next (not pictured).

multiple trading desksThe visual analytics capabilities needed to optimize the cross-functional and collaborative nature of this example are:

  • Building data apps that follow business processes (workflow);
  • Alerting and scheduling based on real-time data;
  • Securely sharing models with internal and external constituents.


Arcadia Data is a platform with the enhanced capabilities described above. Arcadia Data unifies data discovery, business intelligence, and real-time visualization in a single, integrated analytics platform. It enables users to directly leverage big data capabilities (it runs natively in Apache Hadoop clusters, the cloud, or other modern data platforms) through an intuitive and self-service interface. Referring to the examples above, a team can explore billions of rows of highly varied data, join views into interactive packages, and share that package with relevant colleagues to solve problems in a timely, secure, and collaborative way.

This blog was first posted to TabbFORUM, an open community that provides a platform for capital markets professionals to share their ideas and thought leadership with their peers.
[1] Basel Committee on Banking Supervision, Explanatory note on the revised minimum capital requirements for market risk, January 2016, Page 1.
[3] International Swaps and Derivatives Association (ISDA), the Global Financial Markets Association

(GFMA) and the International Institute of Finance (IIF), Industry FRTB QIS Analysis, October 2015.

[4] This is related to Profit and loss (P&L) attribution and Backtesting.  See BCBS explanatory note, p. 5.

Related Posts