Question 1

A hedge fund portfolio has an expected return of 0.1 percent per day and a 5 percent
probability 1-day value at risk (VaR) of $909. Which of the following statement is the best
descriptor of this information?
A) The maximum daily loss on the portfolio is $909.
B) The minimum daily loss on the portfolio is $909.
C) The minimum loss for the worst 5% of the days is $909.
D) The portfolio will earn more than $909 only 5% of the time.
Explanation
By definition, VaR is the minimum loss for the worst 5% of the days or the maximum 1-day
loss 95% of days. A minimum or maximum daily loss on the portfolio of $909 does not
incorporate the alpha (probability). Alternatively, VaR can be stated in terms of confidence,
e.g. in this case you could say you are 95% confident the one-day VaR will not exceed $909
Question 2
You wish to estimate VaR using a local valuation method. Which of the following are
methods you might use?
1. Historical simulation.
II. The delta-normal valuation method.
III. Monte Carlo simulation.
IV. The grid Monte Carlo approach.
A) I and II only.
B) III and IV only.
C) I only.
D) II only.
Explanation
Valuation and Risk Models Local valuation methods measure portfolio risk by valuing the
assets at one point in time, then making adjustments to relevant risk factors that are expected
to cause changes in the overall portfolio value. The delta-normal valuation method is an
example of a local valuation method.
Question 3
All of the following are appropriate methods for addressing return aggregation in volatility
forecasting methods except:
A)the historical standard deviation approach creates a variance-covariance matrix A) that is
estimated under the asasumption that all asset returns are normally distributed.
B) for well-diversified portfolios, the strong law of large numbers is required to estimate the
volatility of the vector of aggregated returns.
C) the historical simulation approach weights returns based on market values today,
regardless of the actual allocation of positions K days ago.
D) the EWMA approach creates a variance-covariance matrix that is estimated under
the assumption that volatility is constant over time
Explanation
Both the EWMA and the historical standard deviation approach create variance-covariance
matrices that are estimated under the assumption that all asset returns are normally
distributed. A major disadvantage of this approach is the number of calculations required to
estimate VaR.
Question 4
The difference between a Monte Carlo simulation and a historical simulation is that a
historical simulation uses randomly selected variables from past distributions, while a Monte
Carlo simulation:
A) uses a computer to generate random variables.
B) projects variables based on a priori principles.
C) uses variables based on roulette odds.
D) uses randomly selected variables from future distributions.
Explanation
A Monte Carlo simulation uses a computer to generate random variables from specified
distributions.
Question 5
The historical standard deviation approach differs from the EWMA and GARCH approaches
for estimating conditional volatility, because it:
A) places a lower weight on more recent data.
B) is a parametric method.
C) uses recent historical data.
D) applies a set of weights to past squared returns.
Explanation
All three methods are parametric, use historical data, and apply weights to past squared
returns. The historical standard deviation approach weighs all returns in the estimation
window equally. The RiskMetricsTM and GARCH approaches are exponential smoothing
approaches that place a higher weight on more recent data.
Question 6
The VaR measure obtained from simulating data based on assumptions concerning the return
distributions is called:
A) Monte Carlo VaR.
B) Kurtotic VaR.
C) Stochastic VaR.
D) Prospective VaR.
Explanation
Monte Carlo VaR uses data generated from a simulation procedure.
Question 7
The Westover Fund is a portfolio consisting of 42 percent fixed income investments and 58
percent equity investments. The manager of the Westover Fund recently estimated that the
annual VaR(5 percent), assuming a 250-day year, for the entire portfolio was $1,367,000
based on the portfolio's market value of $12,428,000 and a correlation coefficient between
stocks and bonds of zero. If the annual loss in the equity position is only expected to exceed
$1,153,000 5 percent of the time, then the daily expected loss in the bond position that will be
exceeded 5 percent of the time is closest to:
A) $46,445.
B) $72,623.
C) $21,163.
D) $55,171.
Explanation
Begin by using the formula for dollar portfolio VaR to compute the annual VaR (5%) for the
bond position:
VaR2 portfolio= VaR2Stocks + VaR2Bonds + 2VaRstocks VaRBonds PStocks, Bonds
(1,367,000)2 = (1,153,000)2+ VaR2Bonds+ 2(1,153,000)VaRBonds(0)
VaRBonds = [(1,367,000)2-(1,153,000)2 )0.5 = 734,357
Next convert the annual $VaRBonds to daily $VaRBonds:
734,357/(250)0.5 = 46,445
Question 8
If the one-day value at risk (VaR) of a portfolio is $50,000 at a 95% probability level, this
means that we should expect that in one day out of:
A) 20 days, the portfolio will decline by $50,000 or more.
B) 95 days, the portfolio will lose $50,000.
C) 95 days, the portfolio will increase by $50,000 or more.
D) 20 days, the portfolio will decline by $50,000 or less.
Explanation
A 95% one-day portfolio value at risk (VaR) of $50,000 means that in 5 out of 100 (or one
out of 20) days, the value of the portfolio will experience a loss of $50,000 or more.
Question 9
Consider the following EWMA models that are used to estimate daily return volatility. Which
model's volatility estimates will have the most day-to-day volatility, and which model will be
the slowest to respond to new data, respectively?
Model 1: σn2 = 0.04µn-12 + 0.96 σn-12
Model 2: σn2 = 0.02µn-12 + 0.98 σn-12
Model 3: σn2 = 0.2µn-12 + 0.80 σn-12
Model 4: σn2 = 0.10µn-12 + 0.90 σn-12
Greatest day-to-day volatility Slowest to respond to new data
A) Model 3 Model 2
B) Model 1 Model 4
C) Model 2 Model 2
D) Model 2 Model 3
Explanation
The form of the basic EWMA model is σ n2 = (1-λ) µn-12 + λ σn-12 where λ is the weight on the
previous volatility estimate. EWMA models with a low value for λ (Model 3) will put more
weight on the previous day's return and will lead to volatility estimates that in themselves are
highly volatile from day to day. EWMA models with a high value for λ (close to 1, such as
Model 2) will put less weight on the previous day's return, and the model will respond more
slowly to new data.
Question 10
Which of the following statements about value at risk (VaR) is true?
A) VaR is independent of probability level.
B) VaR increases with lower significance levels.
C) VaR is not dependent on the choice of holding period.
D) VaR decreases with longer holding periods.
Explanation
VaR measures the amount of loss in the left tail of the distribution and increases with lower
significance levels. VaR actually increases with increases in holding period.
Question 11
An insurance company currently has a security portfolio with a market value of $243 million.
The daily returns on the company's portfolio are normally distributed with a standard
deviation of 1.4%. Using the table below, determine which of the following statements are
true.
Zcritical
Alpha One-tailed Two-tailed
10% 1.28 1.65
2% 2.06 2.32
I. One-day VaR (1%) for the portfolio on a percentage basis is equal to 3.25%.
II. One-day VaR (10%) for the portfolio on a dollar basis is equal to $5.61 million.
III. |One-day VaR (6%) | > |one-day VaR(10%)|
A) I and III only.
B) I, II, and III.
C) I only.
D) II and III only.
Explanation
To find the appropriate Zcritical value for the VaR(1%), use the two-tailed value from the table
corresponding to an alpha level of 2%. Under a two-tailed test, half the alpha probability lies
in the left tail and half in the right tail. Thus the Z critical 2.32 is appropriate for VAR(1%). For
VaR (10%), the table gives the one-tail Z eritical value of 1.28. Calculate the percent and dollar
VaR measures as follows:
VaR1% = z1% x σ = 2.32 x 0.014 = 0.03248 ≈ 3.25%
VaR10% = z10% x σ x portfolio value = 1.28 x 0.014 x $243 million= = $4.35 million
Thus, Statement I is correct and Statement II is incorrect. For Statement III, recall that as the
probability in the lower tail decreases (i.e., from 10% to 6% ), the VaR measure increases.
Thus, Statement III is correct.
Question 12
Which of the following are advantages of nonparametric methods compared to parametric
methods for quantifying volatility?
I. Nonparametric models require assumptions regarding the entire distribution of returns.
II. Data is used more efficiently with nonparametric methods than with parametric methods.
III. Fat tails, skewness, and other deviations from some assumed distribution are no longer a
concern in the estimation process for nonparametric methods.
IV. Multivariate density estimation (MDE) allows for weights to vary based on how relevant
the data is to the current market environment by weighting the most recent data more heavily.
A) I and III.
B) III and IV.
C) I and II.
D) III only.
Explanation
Fat tails, skewness, and other deviations from some assumed distribution are no longer a
concern in the estimation process for nonparametric methods. The other statements are
incorrect for the following reasons:
⚫ Nonparametric models do not require assumptions regarding the entire distribution of
returns.
⚫ Data is used more efficiently with parametric methods than nonparametric methods.
⚫ MDE allows for weights to vary based on how relevant the data is to the current market
environment regardless of the timing of the most relevant data. MDE is also very flexible in
introducing dependence on state variables.
Question 13
Alto Steel's pension plan has $250 million in assets with an expected return of 12 percent.
The last thirty monthly returns are given below.
21.84% -21.50% 31.76% 8.88% 2.54% 17.44%
6.97% 10.00% 2.71% 35.66% 31.07% 18.56%
9.82% -7.94% -0.78% 12.57% 11.77% 8.47%
2.99% 14.35% 14.20% 9.81% 11.03% 22.25%
9.68% 19.55% 8.53% 39.45% 36.15% 10.97%
What is the 10 percent monthly probability VaR for Alto's pension plan?
A) $36,125,850.
B) $3,000,000.
C) $1,200,000.
D) $1,950,000.
Explanation
Sorted monthly returns (from low to high, in columns) are as follows:
-21.50% 2.99% 9.68% 11.03% 17.44% 31.07%
-7.94% 6.97% 9.81% 11.77% 18.56% 31.76%
-0.78% 8.47% 9.82% 12.57% 19.55% 35.66%
2.54% 8.53% 10.00% 14.20% 21.84% 36.15%
2.71% 8.88% 10.97% 14.35% 22.25% 39.45%
The 10% lowest return is the 3rd value (3/30 = 0.10), which is -0.78%.
Therefore 10% VaR for the portfolio = 0.0078 x 250,000,000 = 1,950,000
Question 14
The accuracy of a value at risk (VaR) measure:
A) is one minus the probability level.
B) is included in the statistic.
C) can only be ascertained after the fact.
D) is complete because the process is deterministic.
Explanation
This is a weakness of VaR. The reliability can only be known after some time has passed to
see if the number and size of the losses is congruent with the VaR measure.
Question 15
Which of the following statements about value at risk (VaR) is true?
A) VaR decreases with lower confidence level.
B) VaR decreases with longer holding periods.
C) VaR is not dependent on the choice of holding period.
D) VaR is independent of probability level.
Explanation
Valuation and Risk Models VaR measures the amount of loss in the left tail of the
distribution and increases with lower probability levels. Conversely VaR decreases with
lower confidence levels (which is 1 minus the probability level). VaR actually increases with
increases in holding period.
Question 16
Annual volatility: σ = 20.0%
Annual risk-free = 6.0%
rate
Exercise price (X) =24
Time to maturity =3 months
Stock price, S $21.00 $22.00 $23.00 $24.00 $24.75 $25.00
Value of call, C $0.13 $0.32 $0.64 $1.14 $1.62 $1.80

% Decrease in S -16.00% -12.00% -8.00% -4.00% -1.00%

% Decrease in C -92.83% -82.48% -64.15% -36.56% -9.91%
Delta (∆C%/∆S%) 5.80 6.87 8.02 9.14 9.91
Alton Richard is a risk manager for a financial services conglomerate. Richard generally
calculates the VaR of the company's equity portfolio on a daily basis, but has been asked to
estimate the VaR on a weekly basis assuming five trading days in a week. If the equity
portfolio has a daily standard deviation of returns equal to 0.65% and the portfolio value is $2
million, the weekly dollar VaR (5%) is closest to:
A) $21,450.
B) $47,964.
C) $107,250.
D) $29,100.
Explanation
The weekly VaR is:
2 million x 1.65 x 0.0065 × √5 = $47,964
Question 17
How many of the following statements about VaR methodologies is (are) true?
I. The parametric approach is typically defined by the calculation of the distribution mean
and variance.
II. The nonparametric approach has the advantage of no required asset distribution.
III. The implied-volatility based approach estimates volatility using current market prices.
IV. The GARCH approach is a parametric model that uses time varying weights on historic
returns to calculate distribution parameters.
A) All statements are true.
B) Three statements are true.
C) One statement is true.
D) Two statements are true.
Explanation
All of the statements are true.
Question 18
The most important way in which the Monte Carlo approach to estimating operational VaR
differs from the historical method and variance-covariance method is:
A) its heavy dependence on historical data.
B) it involves repeatedly shocking a model of risk data to produce a range of potential
losses.
C) its inability to account for non-linear risk structures.
D) its computational simplicity.
Explanation
The Monte Carlo approach uses simulation techniques, repeatedly shocking a model of loss
data in order to produce a range of potential losses. It is more computationally intensive than
either the historical or variance-covariance approaches. The model used can account for
nonlinear risk structures and need not be limited by historical data.
Communities Bank has a $17 million par position in a bond with the following
characteristics:
- The bond is a 7-year, zero-coupon bond.
- The market value is $12,358,674.
- The bond is trading at a yield to maturity of 4.6%.
- The historical mean change in daily yield is 0.0%
- The standard deviation of the position is 1%.
Question 19-20
The one-day VaR for this bond at the 95% confidence level is closest to:
A) $105,257.
B) $339,487.
C) $260,654.
D) $203,918.
Explanation
VaR is the market value of the position times the price volatility of the position times the
confidence level, which in this case equals:
($12,358,674) x (0.01) x (1.65) = $203,918
Question 20 - 20
The 10-day VaR on this bond is closest to:
A) $866,111.
B) $644,845.
C) $736,487.
D) $487,698.
Explanation
The VaR is calculated as the dally earnings at risk times the square root of days desired,
which is 10. The calculation generates:
($203,918) (√10) = $644,845
Question 21
All of the following are examples of why returns distributions can deviate from the normal
distribution except the distributions:
A) are fat tailed.
B) are skewed.
C) are symmetrical.
D) have unstable parameters.
Explanation
Examples of common deviations from the normal distribution are fat tails and skewed and/or
unstable parameters. The normal distribution is symmetrical.
Question 22
Which of the following statements regarding volatility in VaR models are true?
I. The RiskMetricsTM approach is very similar to the GARCH model.
II. The historical standard deviation approach creates a variance-covariance matrix that is
estimated under the assumption that all asset returns are normally distributed.
III. The parametric approach typically assumes asset returns are normally or lognormally
distributed with constant volatility.
IV. Exponential smoothing methods and the historical standard deviation methods both apply
a set of weights to recent past squared returns.
A) I, III, and IV.
B) I, II, and III.
C) I, II, and IV.
D) II, III, and IV.
Explanation
The third statement is false. The parametric approach typically assumes asset returns are
normally or lognormally distributed with time-varying volatility. The RiskMetrics TM approach
is actually a special case of the GARCH model. Both the exponential and historical standard
deviation approaches create a variance-covariance matrix that is estimated under the
assumption that all asset returns are normally distributed. Exponential smoothing methods
and the historical standard deviation methods both apply a set of weights to recent past
squared returns. The difference is that in the historical standard deviation method all weights
are equal whereas more recent returns are weighted more heavily in exponential methods.
Question 23
Which of the following statements comparing Monte Carlo VaR and historical VaR is most
accurate?
A) Both are parametric approaches, but historical VaR uses a regression on past data while
Monte Carlo VaR uses Kalman filtering to create forward looking VaR estimates.
B) Both compute VaR from percentiles from a given set of observed returns, but Monte Carlo
VaR uses realized returns and historical VaR uses hypothetical returns.
C) Both are parametric approaches, but Monte Carlo VaR uses fewer inputs into the model
than historical VaR.
D) Both compute VaR from percentiles from a given set of observed returns, but
historical VaR uses realized returns and Monte Carlo VaR uses hypothetical returns.
Explanation
Historical VaR uses historical realized returns, and Monte Carlo VaR uses returns generated
from a hypothetical model, which requires a significant number of inputs. Neither historical
nor Monte Carlo VaR is a parametric approach.
Question 24
Which of the following statements regarding fat-tail distributions is (are) true?
A fat-tailed distribution:
I. most likely results from time-varying volatility for the unconditional distribution.
II. has a lower probability mass around one standard deviation from the mean than a normal
distribution.
III. has a lower probability mass around the mean than a normal distribution.
IV. most likely results from time-varying means for the conditional distribution.
A) I only.
B) II and IV.
C) I and III.
D) I and II.
Explanation
The most likely explanation for "fat tails" is that the second moment or volatility is time-
varying. For example, volatility changes in interest rates are observed prior to much
anticipated Federal Reserve announcements. Examining a data sample at different points of
time from the full sample could generate fat tails in the unconditional distribution even if the
conditional distributions are normally distributed. The conditional mean is not expected to
deviate over time. The first two moments (mean and variance) of the distributions are similar
for the fat-tailed and normal distribution. However, fat-tailed distributions typically have less
probability mass in the intermediate range, around +/-1 standard deviation, compared to the
normal distribution. Fat-tailed distributions have greater mass in the tails and a greater
probability mass around the mean than the normal distribution.
Question 25
Kiera Reed is a portfolio manager for BCG Investments. Reed manages a $140,000,000
portfolio consisting of 30 percent European stocks and 70 percent U.S. stocks. If the
VaR(1%) of the European stocks is 1.93 percent, or $810,600, the VaR (1%) of U.S. stocks is
2.13 percent, or $2,087,400, and the correlation between European and U.S. stocks is 0.62,
what is the portfolio VaR(1%) on a percentage and dollar basis?
A) 2.07% and $2.67 million.
B) 1.90% and $2.90 million.
C) 2.07% and $2.90 million.
D) 1.90% and $2.67 million.
Explanation
VaR for the portfolio on a percentage and dollar basis is calculated as follows:
%VaRportfolio= √0.32 (0.01932)+0.72 (0.02132) + 2(0.3) (0.7) (0.0193) (0.0213) (0.62)
= √0.00003+0.000022 +0.00011
=√0.0036 = 0.01897= 1.897%≈ 1.90%
$VaRportfolio =√810,6002+2,087,4002+2 (810,600) (2,087,400) (0.62)
=√7,112,448,705,600 = $2,666,917 ≈ $2.67 million
Question 26
Tim Jones is evaluating two mutual funds for an investment of $100,000. Mutual fund A has
$20,000,000 in assets, an annual expected return of 14 percent, and an annual standard
deviation of 19 percent. Mutual fund B has $8,000,000 in assets, an annual expected return of
12 percent, and an annual standard deviation of 16.5 percent. What is the daily value at risk
(VaR) of Jones' portfolio at a 5 percent probability if he invests his money in mutual fund A?
A) $13,344.
B) $1,668.
C) $1,924.
D) $38,480.
Explanation
Daily standard deviation for mutual fund A = 0.19 / √250= 0.012
Daily return 0.14/250 = 0.00056
VaR = Portfolio value [E (R) - z σ] = 100,000 (0.00056- (1.65) (0.012)] = -$1,924
Question 27
One advantage of the Monte Carlo simulation approach over the historical method when
calculating VaR is the simulation approach:
A) equates past performance to future results.
B) takes advantage of the normal distribution.
C) incorporates flexibility in modeling price paths.
D) makes better use of computing power.
Explanation
The Monte Carlo approach allows for whatever relationships the VaR modeler would like to
take into account. It is the most flexible method for generating VaR.
Question 28
A portfolio manager determines that his portfolio has an expected return of $20,000 and a
standard deviation of $45,000. Given a 95 percent confidence level, what is the portfolio's
VaR?
A) $94,250.
B) $43,500.
C) $54,250.
D) $74,250.
Explanation
The expected outcome is $20,000. Given the standard deviation of $45,000 and a z-score of
1.65 (95% confidence level for a one-tailed test), the VaR is -54,250 [=20,000 - 1.65
(45,000)].
Question 29
Valuation and Risk Models Super Hedge fund has $20 million in assets. The total return for
the past 40 months is given below. What is the monthly value at risk (VaR) of the portfolio at
a 5 percent probability level?
Monthly Returns
-22.46% 9.26% -4.69% -20.66% -2.77% 1.17% - 16.11% -6.73%
0.57% 12.56% -18.26% -32.81% 24.15% -34.26% -5.49% -19.76%
-34.75% -12.02% 32.74% -31.35% 13.68% -31.13% 7.07% -33.56%
-20.37% 30.27% 31.09% -3.26% -14.42% 4.75% 15.63% -11.57%
7.23% -20.77% -19.61% -2.42% -30.59% 28.83% -22.25% -10.26%
A) $7,200,000.
B) $9,000,000.
C) $16,725,000.
D) $6,852,000.
Explanation
Sorted monthly returns (from low to high, in columns) are as follows:
-34.75% -31.35% -22.25% -19.61% -11.57% -4.69% 0.57% 6.35%
-34.26% -31.13% -20.77% -18.26% -10.26% -3.26% 0.95% 7.07%
-33.56% -30.59% -20.66% -16.11% -6.73% -2.83% 1.17% 7.23%
-33.16% -23.08% -20.37% -14.42% -6.37% -2.77% 1.58% 8.35%
-32.81% -22.46% -19.76% -12.02% -5.49% -2.42% 4.75% 9.26%
The 5% lowest return is the 2nd value (2/40= 0.05), which is -34.26%
Therefore 5% VaR for the portfolio = 0.3426 x $20,000,000= $6,852,000
Question 30
Many analysts prefer to use Monte Carlo simulation rather than historical simulation because:
A) past distributions cannot address changes in correlations or events that have not
happened before.
B) past data is often proprietary and difficult to obtain.
C) it is much easier to generate the required variables.
D) computers can manipulate theoretical data much more quickly than historical data
Explanation
While the past is often a good predictor of the future, simulations based on past distributions
are limited to reflecting changes and events that actually occurred. Monte Carlo simulation
can be used to model based on parameters that are not limited to past experience.
Question 31
A portfolio comprises 2 stocks: A and B. The correlation of returns of stocks A and B is 0.8.
Based on the information below, compute the portfolio's annual VaR at a 5 percent
probability level.
Stock Value E(R) σ
A $75,000 12.0% 15%
B $25,000 10.8% 10.0%
A) $11,700.
B) $23,491.
C) $13,300.
D) $10,295
Explanation
Weight of stock A = WA=0.75; Weight of stock B = WB = 0.25
Expected Portfolio return = E(Rp) = 0.75(12) + 0.25(10.8) = 11.70%
Portfolio Standard deviation =
Sp = [(WA)2 (SA)2 + (WB)2 (SB)2 + 2(WA) (WB) rAB SA SB]0.5 =
[(0.75)2 (0.15)2 + (0.25)2 (0.10)2+2 (0.75) (0.25) (0.8) (0.15) (0.10)]0.5
= (0.0178)0.5 = 13.33%
VaR= Portfolio value [E (R) - zS]= 100,000 [0.117-(1.65) (0.1333)] = -$10,295
Question 32
Which of the following statements most accurately describes the pitfalls of VaR estimation
methods?
I. The Monte Carlo simulation method is subject to model-risk.
II. The historical simulation method is subject to time-variation risk.
III. The delta-normal method will underestimate the VaR for fat-tailed distributions.
A) I, II and III.
B) II and III.
C) I only.
D) I and II.
Explanation
Statements I, II, and III are correct.
Disadvantages of Monte Carlo simulation include:
- model risk;
- sampling variation if replications are small;
- and computational time and skill.
Disadvantages of the historical simulation method
- include: lack of historical data;
 - use of actual data;
- time variation risk;
- inability to recognize structural change;
- and it may not be able to sufficiently define the distributions tails.
Disadvantages of the delta-normal method include:
- normality assumption results in VaR estimates that understate true VaR for
distributions with fat tails;
- and it isn't able to accurately estimate VaR for portfolios with nonlinear
characteristics (i.e., portfolios with option-like positions).
Question 33
A global portfolio is comprised of European and Emerging market equities. The correlation
of returns for the two sectors is 0.3. Based on the information below, what is the portfolio's
annual value at risk (VaR) at a 5 percent probability level?
Stock Value E(R) σ
European $800,000 9.0% 15.0%
Emerging $200,000 18.0% 25.0%
A) $130,300.
B) $128,280.
C) $230,491.
D) $110,700.
Explanation
Weight of European equities = WA = 0.80; Weight of Emerging = WB = 0.20
Expected Portfolio return = E(Rp) = 0.8(9)+0.2(18)= 10.80%
Portfolio Standard deviation =
σP = [(WA)2 (σA)2 + (WB)2 (σB)2 + 2(WA) (WB) rAB σA σB]0.5 =
[(0.8)2 (0.15)2 + (0.2)2 (0.25)2+2 (0.8) (0.2) (0.3) (0.15) (0.25)]0.5
= (0.0205)0.5 = 14.32%
VaR= Portfolio value [E (R) - zσ]= 1,000,000 [0.108-(1.65) (0.1432)] = -$128,280
Question 34
Which of the following is (are) an advantage(s) of nonparametric methods compared to
parametric methods for quantifying volatility?
I. Nonparametric models require assumptions regarding the entire distribution of returns.
II. Data is used more efficiently with nonparametric methods than parametric methods.
III. Fat tails, skewness and other deviations from some assumed distribution are no longer a
concern in the estimation process for nonparametric methods.
IV. Multivariate density estimation (MDE) allows for weights to vary based on how relevant
the data is to the current market environment by weighting the most recent data more heavily.
A) III only.
B) I and III.
C) I and II.
D) III and IV.
Explanation
Fat tails, skewness, and other deviations from some assumed distribution are no longer
concern in the estimation process for nonparametric methods. The other statements a false for
the following reasons. Nonparametric models do not require assumptions regarding the entire
distribution of returns. Data is used more efficiently with param methods than nonparametric
methods. Multivariate density estimation (MDE) allows weights to vary based on how
relevant the data is to the current market environment, regardless of the timing of the most
relevant data. MDE is also very flexible in introducing dependence on state variables.
Question 35
If a 1-day 95 percent VaR is $5 million, the 250-day 99 percent VaR level would be closest
to:
A) $111.79 million.
B) $55.89 million.
C) $21.00 million.
D) $83.84 million.
Explanation
First it is necessary to adjust for confidence levels (2.326/1.645), then by days (√250). In this
case, ($5 million) (2.326/1.645) (√250) = $111.79 million.
Question 36
Which of the following statements about value at risk (VaR) is true?
A) VaR decreases with lower probability levels.
B) VaR increases with longer holding periods.
C) VaR is independent of probability level.
D) VaR is not dependent on the choice of holding period.
Explanation
VaR measures the amount of loss in the left tail of the distribution. It increases with lower
probability levels and increases in holding period.
Question 37
Which of the common methods of computing value at risk relies on the assumption of
normality?
A) Variance/covariance.
B) Rounding estimation.
C) Historical.
D) Monte Carlo simulation.
Explanation
The variance/covariance method relies on the assumption of normality.
Question 38
Which of the following deviations from normality always leads to underestimating the
distribution variance?
I. Higher probability of high returns.
II. Higher probability of mean returns.
III. The mean of the distribution is conditional on the economic environment.
IV. The variance of the distribution is conditional on the economic environment.
A) III and IV only.
B) II only.
C) I, II, and IV only.
D) III only.
Explanation
Statement II suggests that more returns appear in the center of the distribution, which would
likely lead to an underestimation of variance.
Statement I leads to an overestimate of variance. Statement III suggests no change since it is
not likely, in an efficient market, that conditional means vary enough to make a difference
over time. Statement IV leads to over or under estimates of the variance.
Question 39
A distribution of asset returns that has a significantly higher probability of obtaining large
losses is described as:
A) thin-tailed.
B) left skewed.
C) symmetrical.
D) right skewed.
Explanation
A distribution is left skewed when the distribution is asymmetrical and there is a higher
probability of large negative returns than there is for large positive returns.
Question 40
A large bank currently has a security portfolio with a market value of $145 million. The daily
returns on the bank's portfolio are normally distributed with 80% of the distribution lying
within 1.28 standard deviations above and below the mean and 90% of the distribution lying
within 1.65 standard deviations above and below the mean. Assuming the standard deviation
of the bank's portfolio returns is 1.2%, calculate the VaR (5%) on a one-day basis.
A) $2.04 million.
B) cannot be determined from information given.
C) $2.23 million.
D) $2.87 million.
Explanation
VaR (5%)= z5% x σ x portfolio value = 1.65 x 0.012 x $145 million= $2.871 million
Question 42
Value at risk (VaR) is a benchmark associated with a given probability. The actual loss:
A) will have an inverse relationship with VaR.
B) cannot exceed this amount.
C) may be much greater.
D) is expected to be the average of the expected return of the portfolio and VaR.
Explanation
VaR is a benchmark that gives an estimate of what magnitude of loss would not be unusual.
The actual loss for any given time period can be much greater.
Question 43
If a 10-day VaR is $15,000,000, the 250-day VaR, assuming no change in confidence level,
would be:
A) $237,000,000.
B) $7,500,000.
C) $75,000,000.
D) $23,700,000.
Explanation
Just back out the 1-day VaR by dividing by the square root of 10 and then multiply by the
square root of 250 to get the 250-day VaR.
$15,000,000 x √250/√10 =$75,000,000
Question 44
A $2 million balanced portfolio is comprised of 40 percent stocks and 60 percent
intermediate bonds. For the next year, the expected return on the stock component is 9
percent and the expected return on the bond component is 6 percent. The standard deviation
of the stock component is 18 percent and the standard deviation of the bond component is 8
percent. What is the annual VaR for the portfolio at a 1 percent probability level if the
correlation between the stock and the bond component is 0.25?
A) $152,250.
B) $126,768.
C) $149,500.
D) $303,360.
Explanation
Weight of Stock= WS = 0.40; Weight of Bonds= WB = 0.60
Expected Portfolio return = E(Rp) = 0.40(9)+0.60(6)= 7.20%
Portfolio Standard deviation =
σP = [(WS)2 (σS)2 + (WB)2 (σB)2 + 2(WS) (WB) rSB σS σB]0.5 =
[(0.4)2 (0.18)2 + (0.6)2 (0.08)2+2 (0.4) (0.6) (0.25) (0.18) (0.08)]0.5
= (0.009216)0.5 = 9.6%
VaR= Portfolio value [E(R) - zσ]= 2,000,000 [0.072-(2.33) (0.096)] = -$303,360
Question 45
When comparing a fat-tailed distribution to an otherwise similar normal distribution, the fat-
tailed distribution often has:
A) an equal probability mass close to the mean.
B) a lower probability mass at around one standard deviation.
C) a lower probability mass at more than three standard deviations.
D) a different mean and standard deviation.
Explanation
Fat-tailed distributions typically have less probability mass in the intermediate range, around
+/- one standard deviation, compared to the normal distribution. The first two moments
(mean and variance) of the distributions are similar for the fat- tailed and normal
distributions. Fat-tailed distributions have greater mass in the tails and a greater probability
mass around the mean than the normal distribution.
Question 46
For a $1,000,000 stock portfolio with an expected return of 12 percent and an annual standard
deviation of 15 percent, what is the VaR with 95 percent confidence level?
A) $247,500.
B) $150,000.
C) $127,500.
D) $120,000.
Explanation
VaR = Portfolio Value[E(R) - zσ] = 1,000,000[0.12 - (1.65)(0.15)] = -$127,500
Question 47
The price value of a basis point (PVBP) of a bond portfolio is $45,000. Expected changes in
interest rates over the next year are summarized below:
Change in Interest rates Probability
> +1.50% 1%
+1.00-1.49% 29%
0.00-0.99% 20%
-0.99-0.00% 45%
<-1.00% 5%
What is the value at risk (VaR) for the bond portfolio at a 99 percent confidence level?
A) $2,250,000.
B) $4,500,000.
C) $6,750,000.
D) $7,850,500.
Explanation
At 1% probability level change in interest rates is 1.50% or higher.
Change in Portfolio value for a 150 bps change in rates = 150 x 45000 = 6,750,000
VaR = 6,750,000
Question 48
An investor has 60 percent of his $500,000 portfolio in Value fund and the remaining in
Growth fund. The correlation of returns of the two funds is -0.20. Based on the information
below, what is the portfolio's VaR at a 5 percent probability level?
Fund E(R) σ
 Value 12% 14%
Growth 16% 20%
A) $26,768.
B) $82,368.
C) $17,635.
D) $49,824.
Explanation
Weight of Value Fund WV=0.60; Weight of Growth Fund WG = 0.40
Expected Portfolio return = E(Rp) = 0.60(12) + 0.40(16) = 13.60%
Portfolio Standard deviation =
σP = [(WV)2 (σV)2 + (WG)2 (σG)2 + 2(WV) (WG) rVG σV σG]0.5 =
[(0.6)2 (0.14)2 + (0.4)2 (0.2)2+2 (0.6) (0.4) (-0.2) (0.14) (0.2)]0.5= (0.010768)0.5 = 10.38%
VaR= Portfolio value [E(R) - zσ]= 500,000 [0.1360-(1.65) (0.1038)] = -$17,635
Question 49
Which of the following approaches is the most restrictive regarding the underlying
assumption of the asset return distribution?
A) hybrid.
B) multivariate density estimation.
C) parametric.
D) nonparametric.
Explanation
A parametric model typically assumes asset returns are normally or lognormally distributed
with time-varying volatility. The other approaches do not require assumptions regarding the
underlying asset return distribution.
Question 50
Hugo Nelson is preparing a presentation on the attributes of value at risk. Which of Nelson's
following statements is not correct?
A) VaR can account for the diversified holdings of a financial institution, reducing capital
requirements.
B) VaR( 10%) = $0 indicates a positive dollar return is likely to occur on 90 out of 100 days.
C) VAR(1%) can be interpreted as the number of days that a loss in portfolio value will
exceed 1%.
D) VaR was developed in order to more closely represent the economic capital necessary to
ensure commercial bank solvency.
Explanation
VaR is defined as the dollar or percentage loss in portfolio value that will be exceeded only X
% of the time. VaR (10%) = $0 indicates that there is a 10% probability that on any given day
the dollar loss will be greater than $0. Alternatively, we can say there is a 90% probability
that on any given day the dollar gain will be greater than $0. VaR was developed by
commercial banks to provide a more accurate measure of their economic capital
requirements, taking into account the effects of diversification.
Question 51
Portfolio A has total assets of $14 million and an expected return of 12.50 percent. Historical
VaR of the portfolio at 5 percent probability level is $2,400,000. What is the portfolio's
standard deviation?
A) 17.97%.
B) 15.75%.
C) 14.65%.
D) 12.50%.
Explanation
VaR = Portfolio value [E(R) - zσ]
-2,400,000= 14,000,000 [0.125-(1.65) (X)]
X = 17.97%
Note that VaR value is always negative.
Question 52
Which of the following are true about the EWMA, GARCH, and historical standard deviation
approaches to estimate conditional volatility?
I. EWMA and historical standard deviation assume equal weights on all observations.
II. EWMA and GARCH are parametric models: historical standard deviation is not.
III. A decreasing λ suggests a higher relative weight on the most recent data for exponential
smoothing models.
IV. The most recent weight for GARCH exceeds the most recent weight for historical
standard deviation, assuming the same high number of observations.
A) III and IV only.
B) II, III, and IV only.
C) I, II, and IV only.
D) II and III only.
Explanation
EWMA does not assign equal weights across observations. Historical standard deviation is a
parametric model.
Question 53
Which value at risk methodology is most subject to model risk?
A) Variance/covariance.
B) Parametric.
C) Historical.
D) Monte Carlo simulation.
Explanation
Monte Carlo simulation is subject to model risk.
Question 54
A portfolio comprises 2 stocks: A and B. The correlation of returns of stocks A and B is 0.4.
Based on the information below, what is the portfolio's value-at-risk (VaR) at a 5 percent
probability level?
Stock Value E(R) σ
A $85,000 15.0% 18.0%
B $15,000 12.0% 10.0%
A) $11,784.
B) $23,491.
C) $13,300.
D) $1,410.
Explanation
Weight of stock A = WA=0.85; Weight of stock B = WB = 0.15
Expected Portfolio return = E(Rp) = 0.85(15) + 0.15(12) = 14.55%
Portfolio Standard deviation =
Sp = [(WA)2 (SA)2 + (WB)2 (SB)2 + 2(WA) (WB) rAB SA SB]0.5 =
[(0.85)2 (0.18)2 + (0.15)2 (0.10)2+2 (0.85) (0.15) (0.4) (0.18) (0.10)]0.5
= (0.02547)0.5 = 15.96%
VaR= Portfolio value [E (R) - zS]= 100,000 [0.1455-(1.65) (0.1596)] = -$11,784

Question 55
When would a Monte Carlo simulation be preferable to a historical simulation?
A) There is only a small amount of historical data.
B) Insufficient computer capacity.
C) Historical data does not produce favorable results.
D) A large amount of historical data is available.
Explanation
Historical simulation is most applicable if there is a large sample of past returns to draw
from. The computer capacity necessary for each is about the same, and certainly the
occurrence of unfavorable results is no reason to reject historical simulation.
Question 56
RiskMetrics uses the following value for the decay factor of daily data:
A) 0.97.
B) 0.95.
C) 0.92.
D) 0.94.
Explanation
RiskMetrics uses a decay factor of 0.94 for daily data and 0.97 for monthly data.
Question 57
If the expected change in a fixed income portfolio is $520,000 and the standard deviation of
the estimated change in the portfolio is $2,275,500, the 95 percent value at risk (VaR) for this
portfolio is closest to:
A) $3,743,197.50
B) $4,598,597.50
C) $3,223,197.50
D) $855,400.00
Explanation
VaR for this portfolio would be:
[$520,000 - 1.645($2,275,500)] = $3,223,197.50
Question 58
A regime-switching volatility model of interest rates would assume all of the following
except:
A) interest rates are conditionally normally distributed.
B) the regime determines whether the volatility of interest rates is high or low.
C) the unconditional distribution of interest rates is normally distributed.
D) the mean is constant.
Explanation
A regime-switching volatility model assumes different market regimes exist with high or low
volatility. The mean is assumed constant, but the volatility depends on the regime.
Conditional on the fact that interest rates are drawn from one regime, the distribution is
normally distributed. If interest rates are drawn from more than one regime, this
unconditional distribution need not be normally distributed.
Question 59
On December 31, 2006, Portfolio A had a market value of $2,520,000. The historical
standard deviation of daily returns was 1.7%. Assuming that portfolio A is normally
distributed, calculate the daily VaR (2.5%) on a dollar basis and state its interpretation. Daily
VaR(2.5%) is equal to:
A) $70,686, implying that daily portfolio losses will only exceed this amount 2.5% of the
time.
B) $70,686, implying that daily portfolio losses will fall short of this amount 2.5% of the
time.
C) $83,966, implying that daily portfolio losses will only exceed this amount 2.5% of the
time.
D) $83,966, implying that daily portfolio losses will fall short of this amount 2.5% of the
time.
Explanation
VaR (2.5%) Percentage Basis = Z × σ = 1.96(0.017)=0.03332 = 3.332%
VaR (2.5%)Dollar Basis = VaR (2.5%) percentage Basis x portfolio value = 0.03332 x $2,520,000 =
$83,966
The appropriate interpretation is that on any given day, there is a 2.5% chance that the
portfolio will experience a loss greater than $83,996. Alternatively, we can state that there is
a 97.5% chance that on any given day, the observed loss will be less than $83,996.
Question 60
Derivation Inc. has a portfolio of $100 MM. The expected return over one year is 6 percent,
with a standard deviation of 8 percent. What is the VaR for this portfolio at the 99 percent
confidence level?
A) $12.1 MM.
B) $7.2 MM.
C) $12.6 MM.
D) $2.0 MM.
Explanation
VaR= $100 MM [0.06- (2.326) (0.08)] = $12.608 MM
Question 61
The price value of a basis point (PVBP) of a $20 million bond portfolio is $25,000. Interest
rate changes over the next one year are summarized below:
Change in Interest rates Probability
> +2.50% 1%
+2.00-2.49% 4%
0.00-1.99% 50%
-0.99-0.00% 40%
<-1.00% 5%

Compute VaR for the bond portfolio at 95 percent confidence level.

A) $5,000,000.
B) $2,750,000.
C) $2,500,000.
D) $12,500.
Explanation
At 5% probability level change in interest rates is 2.00% or higher.
Change in Portfolio value for 200 bps change in interest rate = 200 x $25,000
VaR = $5,000,000
Question 62
The minimum amount of money that one could expect to lose with a given probability over a
specific period of time is the definition of:
A) the coefficient of variation.
B) value at risk (VaR).
C) the hedge ratio.
D) delta.
Explanation
This is an often-used definition of VaR.