26. As the newly appointed head of operational risk for a large international bank, you must evaluate the company's current approach to estimating the firm-wide operational loss distribution. The bank's current approach is a bottoms-up process in which for each trading desk the operational loss severity distribution is estimated by fitting historical loss magnitude data to a Weibull distribution and the operational loss frequency distribution is estimated by fitting historical loss timing data to a Poisson distribution. Each trading desk's operational loss distribution is then estimated by aggregating the frequency and severity distributions using convolution. Finally, the firm-wide operational loss distribution is estimated using a copula function generated through Monte Carlo simulation. In evaluating this process, which of the following assumptions implied by the current approach will require further investigation?

I. The independence of operational loss events of each particular trading desk.

II. The independence of the frequency of operational loss events and the severity of operational loss events of each particular trading desk.

III. The independence of operational loss events between trading desks.

IV. The reliability and sufficiency of historical loss data for each trading desk.

A. I, II, III and IV

B. I, II and IV

C. II, III and IV

D. I and III

27. As the inexperienced global head of operational risk for DEF Financial Services, you are trying to make a decision about how to quantify your firm's operational risk exposure. You've decided the best combination of methodologies to use would be factor-based models and the capital asset pricing model. Now, 6 months after you've implemented your approach you've identified some major limitations in your decision on the methodologies to use. Which of the following best describes those limitations?

I. You are unable to reliably predict an operational risk event on a detailed level.

II. Collecting and aggregating consistent data across the firm is challenging.

III. You are unable to capture interdependencies between areas of your firm.

IV. You are unable to reliably forecast general trends in operational risk events.

A. I and III

B. II and IV

C. I, II and III

D. II, III and IV

28. Capital is used to protect the bank from which of the following risks:

A. Risks with an extreme catastrophic financial impact

B. High frequency low loss events

C. Low frequency risks with significant severe financial impact

D. High frequency uncorrelated events

29. According to the Basel Committee which of the options below is NOT a qualitative standard that a bank must meet before it is permitted to use the Advanced Measurement Approach (AMA) for operational risk capital:

A. Internal and/or external regulators must perform regular reviews of the operational risk management processes and measurement systems. This review must include both the activities of the business units and of the independent operational risk function

B. There must be regular reporting of operational risk exposures and loss experiences to business unit management, senior management and to the board of directors

C. The bank's internal operational risk measurement system should not be integrated into the day-to-day risk management processes of the bank but should provide a general overview of the operational risks involved in the processes and operations

D. The bank must have an independent operational risk management function that is responsible for the design and implementation of the bank's operational risk framework.

30. Which of the following is a weakness of the top-down approach to measuring operational risk?

A. It fails to consider historical information

B. You cannot use earnings volatility as an indicator of risk potential in this approach

C. Information on specific sources of risk is not provided

D. It is based on the specific mapping of business units, and not the overall organization

26. As the newly appointed head of operational risk for a large international bank, you must evaluate the company's current approach to estimating the firm-wide operational loss distribution. The bank's current approach is a bottoms-up process in which for each trading desk the operational loss severity distribution is estimated by fitting historical loss magnitude data to a Weibull distribution and the operational loss frequency distribution is estimated by fitting historical loss timing data to a Poisson distribution. Each trading desk's operational loss distribution is then estimated by aggregating the frequency and severity distributions using convolution. Finally, the firm-wide operational loss distribution is estimated using a copula function generated through ffice:smarttags" />Monte Carlo simulation. In evaluating this process, which of the following assumptions implied by the current approach will require further investigation?

I. The independence of operational loss events of each particular trading desk.

II. The independence of the frequency of operational loss events and the severity of operational loss events of each particular trading desk.

III. The independence of operational loss events between trading desks.

IV. The reliability and sufficiency of historical loss data for each trading desk.

A. I, II, III and IV

B. I, II and IV

C. II, III and IV

D. I and III

Correct answer is B

I.  Correct. The frequency of operational loss events is typically assumed to follow a Poisson distribution, but only with the caveat that actual loss events tend to be more correlated than those represented by the theoretical distribution, and investigating whether loss event correlations are too great to assume a Poisson distribution would be prudent.fficeffice" />

II.  Correct.  Using convolution to aggregate severity and frequency distributions assumes independence, so one would want to check that these distributions are in fact uncorrelated.

III.  Incorrect.  Since the current approach uses a copula function to estimate the firm-wide operational loss distribution, the current approach does not assume independence of operational loss events between trading desks.

IV.  Correct.  Estimating loss distributions from historical data assumes that the data is both reliable and sufficient to generate an accurate estimate, and investigating the quality of the historical data would be necessary.

Reference: Allen, Boudoukh, and Saunders, Understanding Market, Credit and Operational Risk: A Value-at-Risk Approach.

27. As the inexperienced global head of operational risk for DEF Financial Services, you are trying to make a decision about how to quantify your firm's operational risk exposure. You've decided the best combination of methodologies to use would be factor-based models and the capital asset pricing model. Now, 6 months after you've implemented your approach you've identified some major limitations in your decision on the methodologies to use. Which of the following best describes those limitations?

I. You are unable to reliably predict an operational risk event on a detailed level.

II. Collecting and aggregating consistent data across the firm is challenging.

III. You are unable to capture interdependencies between areas of your firm.

IV. You are unable to reliably forecast general trends in operational risk events.

A. I and III

B. II and IV

C. I, II and III

D. II, III and IV

Correct answer is C

I.  Correct. Factor-based models and the capital asset pricing model both allow for forecasting general trends but are not reliable for predicting on a detailed level.

II.  Correct. Both methodologies require sufficient data to implement and collecting and aggregating consistent data across any global firm is challenging.

III.  Correct.  A disadvantage both methods is their inability to capture overlaps between areas of a firm

IV.  Incorrect.  Factor-based models and the capital asset pricing model both allow for forecasting general trends but are not reliable for predicting on a detailed level.

28. Capital is used to protect the bank from which of the following risks:

A. Risks with an extreme catastrophic financial impact

B. High frequency low loss events

C. Low frequency risks with significant severe financial impact

D. High frequency uncorrelated events

Correct answer is C

Management knows that in the course of ordinary business, certain operations will fail.  There will be a normal amount of operational loss that a business is willing to absorb as a cost of doing business.  These failures are explicitly budgeted for in the annual business plan and are covered by the price of goods and services.  Capital, on the other hand, will be used to cover low frequency events with significant severe financial impact.  Catastrophic losses are the most extreme but also the rarest types of losses.  Banks will try to find insurance to hedge catastrophic losses because capital will not protect bank from these risk.

29. According to the Basel Committee which of the options below is NOT a qualitative standard that a bank must meet before it is permitted to use the Advanced Measurement Approach (AMA) for operational risk capital:

A. Internal and/or external regulators must perform regular reviews of the operational risk management processes and measurement systems. This review must include both the activities of the business units and of the independent operational risk function

B. There must be regular reporting of operational risk exposures and loss experiences to business unit management, senior management and to the board of directors

C. The bank's internal operational risk measurement system should not be integrated into the day-to-day risk management processes of the bank but should provide a general overview of the operational risks involved in the processes and operations

D. The bank must have an independent operational risk management function that is responsible for the design and implementation of the bank's operational risk framework.

Correct answer is C

'C' is not a qualitative standard. According to the Basel Committee, a "bank's internal operational risk measurement system must be closely integrated into the day-to-day risk management processes of the bank. Its output must be an integral part of the process of monitoring and controlling the bank's operational risk profile."

30. Which of the following is a weakness of the top-down approach to measuring operational risk?

A. It fails to consider historical information

B. You cannot use earnings volatility as an indicator of risk potential in this approach

C. Information on specific sources of risk is not provided

D. It is based on the specific mapping of business units, and not the overall organization

Correct answer is C

The top-down approach does not provide information on specific sources of risk.  It levies an overall cost of operational risk to the entire firm.

A is incorrect.  The top-down approach is based on historical information.

B is incorrect.  Earnings volatility is suggested as an indicator of risk potential for this approach.

C is correct.  Information on specific sources of risk in the organization is not provided.

D is incorrect.  This is a strength of the bottom-up approach, not a weakness of the top-down approach.

Reference: Understanding Market, Credit, and Operational Risk, Allen, Boudoukh, and Saunders, 2004.