AIM 1: Compare and contrast top-down and bottom-up approaches to measuring operational risk.
1、From an operational risk perspective, the risks that are unlikely to jeopardize the future of the firm are:
A) low-frequency, high-severity risks.
B) risks related to business practices.
C) high-frequency, low-severity risks.
D) risks related to internal fraud.
The correct answer is C
Due to their predictable nature and small size, the risks that are unlikely to jeopardize the future of the firm are high-frequency, low-severity events. Note that risks related to fraud and business practices tend to create the largest unexpected losses and can therefore jeopardize the firm.
2、The difference between bottom-up methods for measuring operational risk and top-down methods for measuring operational risk is that bottom-up methods focus on:
A) quantitative rather than qualitative measures of risk.
B) loss causes rather than just loss indicators.
C) loss indicators rather than just loss causes.
D) qualitative rather than quantitative measures of risk.
The correct answer is B
The bottom-up difference is that the focus is on loss causes rather than just loss indicators. Bottom-up operational risk measures may be either quantitative or qualitative.
3、Which of the following approaches to operational risk management can differentiate between high-frequency, low-severity (HFLS) and low-frequency, high-severity (LFHS) events?
A) Top-down approach.
B) Both the top-down approach and bottom-up approach.
C) Bottom-up approach.
D) Neither the top-down approach nor the bottom-up approach.
The correct answer is C
The bottom-up approach analyzes risk in individual processes, which can distinguish between HFLS and LFHS events. The top-down approach to operational risk measurement examines the aggregate impact of internal operational failures by estimating the variance of economic variables that is left unexplained by external macroeconomic factors and does not distinguish between HFLS and LFHS events.
4、The operational risks that will tend to have low expected losses but relatively high unexpected losses are:
A) low-frequency, low-severity risks.
B) low-frequency, high-severity risks.
C) high-frequency, low-severity risks.
D) high-frequency, high-severity risks.
The correct answer is B
Lack of adequate internal data makes the losses associated with low-frequency, high-severity risks the least predictable of operational risks.
5、The process of measuring operational risk may categorize potential losses into groups in terms of:
A) size and frequency.
B) size and style.
C) frequency and style.
D) probability and frequency.
The correct answer is A
The first step in measuring operational risk is to categorize potential losses into groups in terms of size and frequency.
6、The risks that will tend to have high expected losses but relatively low unexpected losses are:
A) high-frequency, high-severity risks.
B) low-frequency, high-severity risks.
C) high-frequency, low-severity risks.
D) low-frequency, low-severity risks.
The correct answer is C
Data availability and frequent occurrence make the losses associated with high-frequency, low-severity risks the most predictable of operational risks.
7、The measure of operational risk in top-down factor models is:
A) R2.
B) operating leverage.
C) estimated factor coefficient.
D) residual variance.
The correct answer is D
factor model explains the variation in the dependent variable (e.g., stock returns, revenue, or expenses) with macroeconomic factors through regression analysis. The R2 measures the proportion of variation explained by the factors. Operational risk is the idiosyncratic variation left unexplained by the factors, or 
, called residual variance. Coefficient estimates measure the sensitivity of the operating variable to changes in the associated macroeconomic factor. Operating leverage is the change in variable costs for a given change in total assets. A measure of operational risk is unexpected changes in operating leverage.
8、Sometimes the debate regarding operational risk focuses more on how to properly define it, rather than how to manage it. Operational risk is commonly defined as any risk that is not classified as:
A) risk-adjusted.
B) an indirect loss.
C) market or credit risk.
D) a direct loss.
The correct answer is C
There is no agreed upon central definition of operational risk. However, a standard or common definition would be any risk not characterized as market or credit risk. There are many definitions, thus the importance of BIS (Bank of International Settlement) having a standardized definition of operational risk.
9、All of the following are components of an effective operational risk management system EXCEPT:
A) calculating and executing accurate settlement payments.
B) processing transactions.
C) projecting default probabilities for counterparties based upon historical experience.
D) collecting data effectively.
The correct answer is C
Projecting default probabilities would address credit risk, not operational risk.
10、Quantifying operational risk is difficult because:
A) the variance of the market is difficult to compute.
B) there are no well-defined procedures or methods.
C) the covariance with the market portfolio is difficult to compute.
D) the time period needed for measurement is very long.
The correct answer is B
Operational risk management is difficult to quantify because there are no well-defined procedures or methods. Other choices are applicable for measuring beta or systematic risk of equity.
11、Operational risk is defined as all of the following EXCEPT:
A) risk of breakdown in normal operations.
B) risk of a swap counterparty not fulfilling their obligation.
C) financial risk that is not caused by market risk.
D) risk of fraud.
The correct answer is B
Risk of a swap counterparty defaulting is called as credit risk. All of the other choices are examples of operational risk.
12、Operational risk sources for a bank include all of the following EXCEPT:
A) employee misappropriation.
B) non-adherence to policy.
C) default on loans obtained by submitting false documents.
D) absence of standards or policies.
The correct answer is C
Default by customer is part of credit risk irrespective of whether the documents were correct or not.
13、Operational risk is defined as all of the following EXCEPT:
A) risk of rate decrease for a utility.
B) financial risk that is not caused by market risk.
C) risk of breakdown in normal operations.
D) risk of inefficient or insufficient systems.
The correct answer is A
Operational risk is risk of losses from failed systems, processes, and people. Risk of rate decrease for a utility is part of market risk.
14、Operational risk is all risk that is NOT:
A) credit or market risk.
B) exchange rate or interest rate risk.
C) credit or interest rate risk.
D) exchange rate or credit risk.
The correct answer is A
Operational risk is also defined as the residual risk – all risk that is not credit or market risk.
15、Operational risk is most poorly defined because:
A) the definition of human risk changes periodically.
B) it is a component of overall market risk.
C) it is difficult to measure accurately.
D) it is poorly correlated with exchange rate risk.
The correct answer is C
Operational risk is the most poorly defined component of a firm’s overall risk because among other things, it is difficult to measure accurately. The human risk component of operational risk is well defined but is also difficult to measure. Operational risk is not a component of market risk and correlation with exchange rate risk is irrelevant.
16、The measurement of which of the following risks is least developed?
A) Credit risk.
B) Market risk.
C) Operational risk.
D) Business risk.
The correct answer is C
The measurement of operational risk is the least developed of the risks facing most firms.
17、Operational risk is defined as all of the following EXCEPT:
A) risk of breakdown in normal operations.
B) financial risk that is not caused by market risk.
C) risk of default by a foreign customer.
D) risk of fraud.
The correct answer is C
Operational risk does not include default risk, which would be considered credit risk.
18、The risks that will have the largest impact on the capital charge for operational risk are:
A) low-frequency, low-severity risks.
B) high-frequency, low-severity risks.
C) high-frequency, high-severity risks.
D) low-frequency, high-severity risks.
The correct answer is D
The risks that will have the most impact on the capital charge for operational risk are the low-frequency, high-severity risks.
AIM 2: List and describe examples of top-down models for measuring operational risk.
1、Residual variance is an estimate of operational risk in which of the following models?
A) Connectivity models.
B) Multi-factor models.
C) Scenario analysis.
D) Reliability models.
The correct answer is B
Residual variance is the proportion of variation in the dependent variable of a regression that is left unexplained by the independent variables. A multi-factor model is a top-down approach to measuring operational risk that uses macroeconomic factors to explain a firm’s stock returns. What is left unexplained by the model is considered aggregate operational risk. Scenario analysis is a top-down approach that does not use regression analysis. Reliability and connectivity models are bottom-up approaches.
2、Which of the following models does NOT use regression analysis to measure operational risk by relating a financial variable to macroeconomic variables?
A) Multi-factor models.
B) Income-based models.
C) Expense-based models.
D) Risk-profiling models.
The correct answer is D
Risk profiling models relate performance indicators to control indicators to identify operational weaknesses. As such, they are not in the class of top-down approaches that relate macroeconomic factors to financial variables. For example, multi-factor models regress stock returns against macroeconomic factors. The residual variance from this regression measures the unexplained variance that is attributable to operational risk. Income-based models and expenses-based models are similar in that they relate measures of income or expenses to macroeconomic factors.
3、In the context of measuring operational risk, operating leverage models:
A) relate changes in normalized expenses to macroeconomic risk variables through regression analysis.
B) measure the change in the relationship between variable costs and total assets.
C) speculate about a set of possible catastrophic events.
D) are relatively subjective.
The correct answer is B
Operating leverage models measure the risk that variable operating costs will increase by more than their historical relation to asset growth would suggest. Although they are a class of expense-based models in some sense, they do not relate expenses to macroeconomic factors. They are relatively objective (not subjective) and do not capture reputational considerations or the opportunity costs. Scenario analysis speculates about a set of possible catastrophic events.
4、Which of the top-down approaches of measuring operational risk focus on LFHS?
A) Risk Profiling Models.
B) Scenario Analysis.
C) Operating Leverage Models.
D) Multifactor Models.
The correct answer is B
Scenario Analysis is unique among top-down approaches focusing on LFHS events that have not occurred yet.
AIM 3: List and describe examples of bottom-up models for measuring operational risk.
1、The primary focus of Extreme Value Theory (EVT) is the notion that:
A) large losses occur more frequently than many assumed distributions suggest.
B) low frequency, high severity losses are more common than losses of more moderate magnitude.
C) the risk of extreme events is defined by the percentile established by a predetermined confidence interval.
D) the possibility of extreme events is sufficiently small that it can safely be ignored.
The correct answer is A
EVT recognizes that the empirical distribution of losses contains “fat tails”. That is, extremely large losses are more frequent than many assumed distributions suggest. It therefore, develops special methods for examining risk in this part of the distribution. Rather than estimating downside risk using the value at risk, say, the 99% confidence interval, EVT estimates the expected value of losses beyond the 99% percentile – an approach the generally yields greater estimates of catastrophic risks.
2、Which of the following models analyze operational risk by deconstructing a process into individual steps and analyzing each step’s risk or each step’s relation to other steps?
I. Causal Networks.
II. Fishbone Analysis.
III. Extreme Value Theory.
IV. Empirical Loss Distributions.
A) II and IV only.
B) I and III only.
C) I, II, III, and IV.
D) I and II only.
The correct answer is D
The question describes process approaches, as opposed to actuarial approaches, to developing bottom-up models for operational risk analysis. Empirical loss distributions examine historical loss data rather than focusing on processes, while extreme value theory focuses on analyzing the tails of a distribution that contain low-frequency, high-severity events. Causal networks, by contrast, dissect a process into sequential steps that can be examined individually. Also a process approach, fishbone analysis is a type of connectivity model that determines the different causes of losses or errors in each step of the process.
3、“Convolution” means:
A) projecting frequency and severity probability distributions into independent losses over a time period.
B) analyzing frequency and severity loss data with regression analysis techniques.
C) reducing loss frequency and severity through a coordinated risk-management program.
D) combining frequency and severity probability distributions into independent losses over a time period.
The correct answer is D
Convolution means combining frequency and severity probability distributions into independent losses over a time period.
4、The two variables that are modeled in probability distribution functions of operational risk are loss:
A) frequency and loss severity.
B) type and loss severity.
C) frequency and loss cause.
D) frequency and loss type.
The correct answer is A
The two variables that are modeled in probability distribution functions of operational risk are loss frequency and loss severity.
5、Referring to the bottom-up models for measuring operational risk, how many of the actuarial models use a standardized distribution to estimate one or several components of operational risk? How many of the actuarial models focus on LFHS?
A) 2; 2.
B) 2; 1.
C) 1; 2.
D) 1; 1.
The correct answer is B
Parametric loss distributions and Extreme Value Theory use standardized distributions and Extreme Value Theory focuses on LFHS.
6、Which of the following characterize extreme value theory (EVT)?
Focuses on catastrophic losses that are more likely than standard distributions suggest.
Uses a different distribution to describe losses in the right tail.
Most commonly uses a lognormal distribution in the tails.
Estimates expected losses beyond an established confidence interval.
A) I and III only.
B) I and IV only.
C) I, II, and IV only.
D) II and III only.
The correct answer is C
EVT incorporates the notion that extreme losses are more likely than standard distribution would suggest. In other words, empirical distributions often have "fatter" tails than assumed distributions. EVT therefore treats the tails of assumed distributions differently than the rest of the distribution by assuming this region follows a different distribution, such as the generalized Pareto distribution (most common). Rather than estimating losses up to, say, the 99th percentile as in value-at-risk (VAR) analysis, EVT estimates the expected value of the losses beyond the 99th percentile in the "fat" tail of the distribution. This approach yields estimates of operational risk that are much larger than standard VAR approaches.
AIM 4: List and describe ways a firm can hedge against catastrophic operational losses.
1、Which of the following help insurance companies to manage the moral hazard problem associated with insuring operational risks?
I. Deductibles.
II. Reinsurance.
III. Co-insurance.
IV. Diversification.
A) I and III only.
B) II and III only.
C) II and IV only.
D) I, II, III, and IV.
The correct answer is A
Moral hazard is the notion that an insured will engage in more risky behavior than would be the case in the absence of insurance. Deductibles and co-insurance features cause the insured to participate in at least a portion of losses incurred by the firm. An insurance policy with a deductible does not cover losses below the deductible amount. A policy with a co-insurance feature does not cover losses above the co-insurance limit. Diversification and reinsurance are techniques insurance companies use to manage other risks.
2、Parametric notes are fixed income instruments with cash flows:
A) determined by parametric distributions.
B) that are triggered by internal risk events, such as fraud.
C) linked to an index of underwriting losses.
D) linked to an external risk event, such as an earthquake.
The correct answer is D
All the bonds described above, except for one, are types of catastrophe bonds. Parametric notes link cash flows to the magnitude of an external risk event, such as hurricane severity in a particular region. Indemnified notes offer the issuing firm debt relief based on internal events, such as a large underwriting loss for an insurance company. Indexed notes provide cash flows related to the value of an independent index, such as a weather index or an insurance underwriting loss index. Bonds with cash flows determined by parametric distributions are quixotic
3、Which of the catastrophe bonds are event based? Which are subject to the moral hazard problem?
A) Parametric; Indemnified.
B) Indemnified; Indemnified, Parametric.
C) Indemnified, Parametric; Indemnified.
D) Indemnified; Indemnified.
The correct answer is C
Indemnified notes are based on internal events and are subject to a moral hazard problem. Parametric notes are based on external events.
AIM 5: Describe the characteristics of catastrophe options and catastrophe bonds.
1、The payoff for CBOT catastrophe options is most like:
A) a long put option.
B) a covered call option.
C) an insurance policy with a deductible and co-insurance feature.
D) a short futures contract.
The correct answer is C
The CBOT catastrophe option is designed as a spread option based on an index of underwriting property losses experienced by a large pool of insurers. Its spread feature combines a long call position with a low exercise price with a short call position at a higher exercise price. As a result, losses less than the lower exercise price and greater than the higher exercise prices are uninsured, creating a net payoff similar to an insurance policy with a deductible and co-insurance feature. Long puts and covered calls do not have spread-like payoffs.
2、The cash flows of catastrophe bonds might be linked to:
I. the firm’s dividend payments.
II. internal loss events.
III. industrywide underwriting losses.
IV. an operational index.
A) II, III, and IV.
B) I only.
C) III and IV only.
D) II and III only.
The correct answer is A
Cat bonds, as they are known, can include cash flows from internal risk events (as in the case of indemnified notes), external risk events (as with parametric notes), or the value of an index (as with indexed notes). The firm’s dividend payment, however, is not a risk event and can be easily manipulated by the firm.
AIM 6: Discuss limitations to operational risk hedging.
1、In general, operational risk management:
A) is objective.
B) can accurately incorporate correlations among risk events.
C) is challenged by the need to identify operational risks.
D) benefits from many databases with strong predictive value.
The correct answer is C
Operational risk management is challenged by its generally subjective nature. Even identifying the potential risk events and the correlations among these risks requires subjective analysis, thereby compromising the accuracy of the analysis. Part of the reason for the high degree of subjectivity is the lack of generally available data that can be used to confidently project future risks.
2、Which of the following are limitations to operational risk hedging?
Identifying operational risks is a highly subjective process.
Distributions of loss frequency and loss severity are unavailable.
Risk correlations are difficult to forecast around extreme events.
A) I only.
B) I and II only.
C) II and III only.
D) I and III only.
The correct answer is D
Identifying and measuring operational risk is highly subjective. Correlations are not necessarily stationary, especially during extreme events.
图片附件: 1.jpg (2009-6-30 15:46, 1.7 KB) / 下载次数 1| 欢迎光临 CFA论坛 (http://forum.theanalystspace.com/) | Powered by Discuz! 7.2 |