Quantitative Analysis

1. Miller funds of funds has invested in various hedge funds. The hedge funds where they have
invested are either equity only hedge funds or fixed income only hedge funds. 60% of the equity
only hedge funds and 40% of the fixed income only hedge funds have provided negative returns
in the recent year. Miller funds of funds have invested in 55 equity only hedge funds and 35
fixed income only hedge funds. The fund manager at Miller randomly selected a hedge fund out
of the hedge funds where they have invested. That hedge fund has provided negative return in
recent year. What is the probability that the hedge fund is an equity only hedge fund?
a) 60.00%
b) 61.11%
c) 70.21%
d) Cannot be determined

2. Which of the following statements is/are least accurate?

(1) If A and B are independent events, then Cov(A,B) = 0
(2) If Cov(A,B) = 0, then A and B are independent events
(3) The diversification benefits are maximum when the events are independent
(4) The diversification benefits are minimum when the events are independent
a) (3) and (4)
b) (4) only
c) (1) and (3)
d) (2), (3) and (4)

3. The joint probability of event A and event B is 0.64. The unconditional probability of event A is
twice the unconditional probability of event B. What is the unconditional probability of the
occurrence of event B given that event A and event B are independent?
a) 21.33%
b) 40.00%
c) 56.57%
d) 80.00%

4. The unconditional probability of an increase in stock price of a company is 0.62. There can be
only three states of economy - good, normal and bad. The probability of good times is 0.20 and
the probability of normal times is 0.50. The probability of the stock price going up given that the
economy is going through good times is 0.85. The probability of the stock price going up given
that the economy is going through bad times is 0.20. What is the probability of the stock price
going up given that the economy is going through normal times?
a) 0.39
b) 0.70
c) 0.78
d) 0.81

©2012 Konvexity All Rights Reserved Page 1

5. Fermion Majorana is contemplating on raising capital for his startup company. He will either use
all capital by using debt or by using equity. The probability of raising capital through equity
depends on the decision of local government to allow FIIs in the market. The conditional
probability of raising capital through equity given that FIIs are allowed to invest is 0.80. The
probability of government allowing FIIs is 0.60. The conditional probability of allowance of FII
given that he has raised capital via equity is 0.70. What is the unconditional probability of raising
capital through equity?
a) 52.50%
b) 68.57%
c) 76.00%
d) 93.33%

6. You want to invest your savings in a mutual fund. You want to choose a mutual fund which has
provided the highest annual rate of return in last 5 years. There are 3 funds to choose from.
Fund A has provided a return of 12.50% per annum. Fund B has provided a rate of return of
12.70% and Fund C has provided a rate of return of 12.30%. Fund A is using arithmetic average
to show the annual return while Fund B and Fund C are using geometric average and harmonic
average respectively. Which of the mutual funds you would definitely not choose if you have to
select only one fund?
a) Fund A
b) Fund B
c) Fund C
d) Fund A and Fund C

7. Which of the following statements is/are least accurate?

(1) If X and Y are random independent variables, then Var(X+Y) = Var(X) + Var(Y)
(2) If X and Y are random independent variables, then Var(X-Y) = Var(X) + Var(Y)
(3) Var(4X + 16) = 16 Var(X)
(4) If X and Y are random independent variables, then Var(3X+2Y) = 3Var(X) + 2Var(Y)
a) (2) and (4)
b) (4) only
c) (3) and (4)
d) (2) and (3)

8. If Cov(X,Y) = 0.84 and the standard deviation of X and Y are 0.9 and 1.2 respectively, what is the
correlation coefficient between 4X+5 and 3Y-9?
a) 0.22
b) 0.72
c) 0.78
d) 0.97

©2012 Konvexity All Rights Reserved Page 2

9. The variance of returns of a portfolio of stock is 0.60 and the variance of returns of S&P 500
futures contract is 0.40. The correlation coefficient between the returns of portfolio and the
returns of S&P 500 futures contract is 0.90. The value of the portfolio is $2 million. What should
be the position in S&P 500 futures contract so that the hedge has minimum variance?
a) Short S&P 500 futures worth $2,204,541
b) Short S&P 500 futures worth $2,700,000
c) Long S&P 500 futures worth $2,700,000
d) Short S&P500 futures worth $1,469,694

10. Which of the following distribution is most likely to have the fattest tail?
a) A leptokurtic distribution with kurtosis = 6
b) A leptokurtic distribution with excess kurtosis = 4
c) A normal distribution
d) A platykurtic distribution

11. It is assumed that the returns of a stock follow continuous uniform distribution. The range of the
distribution is from -30% to 45%. What is the probability that the return of the stock will beat
the S&P 500 market expected return. The expected return from the S&P 500 market is 12%?
a) 44.00%
b) 50.00%
c) 56.00%
d) 73.33%

12. What is the standard deviation of returns of a stock following continuous random distribution if
the range of the distribution is from 40% to -5%?
a) 10.10%
b) 12.99%
c) 15.46%
d) 22.50%

13. The return of a stock is following binomial distribution. Either it can go up or it can go down. The
probability of going up is 60% and the probability of going down is 40%. What is the probability
that the stock will have at least 3 positive days out of 5 trading days. Also calculate the variance
of the returns of the stock for the given period.
a) Probability = 0.3456, Variance = 1.20
b) Probability = 0.3456, Variance = 0.24
c) Probability = 0.6826, Variance = 0.24
d) Probability = 0.6826, Variance = 1.20

14. There is 2% probability that the portfolio of fixed income securities will suffer a loss of more
than 10% in a single day. The number of days having losses exceeding 10% follows Poisson
distribution. What is the probability that there will be 10 days in a particular year? Assume that
the number of trading days in a year to be 250 days.
a) 5.14%
b) 3.78%
c) 1.81%
d) 0.57%

©2012 Konvexity All Rights Reserved Page 3

15. Which of the following statements is least accurate about the lognormal distribution?
a) The lognormal distribution is used to model prices of derivative instruments
b) The lognormal distribution is positively skewed
c) The lognormal distribution is bounded from below by zero
d) The probability density function of a lognormal distribution is given by f(x) = (1/xσ√2π)*e-
0.5{(lnx - µ)/σ}^2

16. What is the probability that the value of a random variable following normal distribution lies
between one standard deviation left of mean to two standard deviation right of mean? F(2) =
0.9772 and F(1) = 0.8413.
a) 13.59%
b) 63.59%
c) 68.26%
d) 81.85%

17. Which of the following statements is least accurate about the students' t-distribution?
a) It is defined by the less number of parameters than a normal distribution
b) It has fatter tails than normal distributions
c) The confidence interval of t-distribution are narrower as compared to normal distribution
d) It is bell shaped probability distribution that is symmetrical about its mean

18. The mean return of a mutual fund is 12.00% per annum. The returns are assumed to follow
normal distribution. The standard deviation of returns is 20% per annum. If you plan to invest in
that mutual fund, what is the probability that you won't lose your money next year? F(0.4) =
0.6554 and F(0.6) = 0.7257.
a) 27.43%
b) 34.46%
c) 54.86%
d) 61.89%

19. The 95% confidence interval for a normally distributed sample of returns of a stock is [-12.50%
to 37.50%]. What is the 99% confidence interval for the same stock? F(0.95) =1.645, F(0.975) =
1.96, F(0.99) = 2.33, F(0.0995) = 2.575.
a) [-22.91%, 47.91%]
b) [-17.22%, 42.22%]
c) [-20.34%, 45.34%]
d) Data insufficient

20. Hasan Sadozye is planning to test a hypothesis on a sample mean of returns of a portfolio. He
fears that the distribution of returns of the portfolio might be nonnormal. So, he takes sample
size of more than 30. The variance of the population of returns of portfolio is known. Which test
should be used by him?
a) Chi-square test
b) t-test
c) z-test
d) Either t-test or z-test

©2012 Konvexity All Rights Reserved Page 4

21. Roger tests a hypothesis on a sample of returns from a mutual fund. He checks whether the
returns are lesser than -12% or not and finds out the test statistic to be 1.80. What should be
the result of the hypothesis test assuming that he is using z-test for the significance level of 5%.
Also calculate the probability of Type I error.
a) Rejection of null hypothesis, 2.5% Type I error probability
b) Failed to reject the null hypothesis, 2.5% type I error probability
c) Rejection of null hypothesis, 5% Type I error probability
d) Failed to reject the null hypothesis, 5% Type I error probability

22. Which of the following equations is/are the examples of non-linear relationships between
dependent and independent variable?
I. E(Yi/Xi) = B0 + B1Xi-2
II. E(Yi/Xi) = B0 + B1 ln(Xi)
III. E(Yi/Xi) = B0 + B12 Xi
IV. E(Yi/Xi) = B0 + (B0/ B1) Xi
a) II and IV
b) III and IV
c) I, II and III
d) I, II, III and IV

23. Vincent is trying to develop a linear regression equation on the impact of change in GDP on the
changes in S&P 500 index. He calculates the following results: Mean of changes in GDP = 7%,
Mean of changes in S&P 500 index = 11.5%, Standard deviation of changes in GDP = 5%,
Standard deviation of changes in S&P 500 index = 20.0%, Covariance between changes in GDP
and changes in S&P 500 index = 0.008295. Calculate the slope coefficient for the regression
equation.
a) 0.0415
b) 0.1659
c) 0.8295
d) 3.318

24. What is the coefficient of determination in the Question no. 23?

a) 0.4554
b) 0.6881
c) 0.8295
d) 0.9108

25. In a linear regression equation between the percentage change in profitability

ratio(independent variable) and the percentage change in stock price(dependent variable), the
slope coefficient is 0.5 and the intercept coefficient is 5%. What is the expected percentage
change in the stock price if the percentage change in the profitability is 18% ?
a) 9.00%
b) 14.00%
c) 36.00%
d) 41.00%

©2012 Konvexity All Rights Reserved Page 5

26. The correlation coefficient between the return of a stock and its beta is 0.524. How much
percentage of the variation in the return of a stock is not explained by its beta assuming that
there is a linear regression equation between the return of a stock and its beta?
a) 27.46%
b) 47.60%
c) 52.40%
d) 72.54%

27. The 95% confidence interval for the estimated slope coefficient of linear regression equation
between the return of an index (dependent variable) and market risk premium is 0.65 to 1.35.
What is the total expected change in the return on the index if the market risk premium changes
from 8% to 10%?
a) 1.30%
b) 2.00%
c) 2.70%
d) Can't be determined

28. The estimated slope coefficient for a linear regression equation is 1.2 with a standard error
equal to 0.65. The sample has 25 observations. Is the estimated slope coefficient significantly
different than zero at 1% level of confidence? What is the value of test statistic?
a) Yes, 1.8462
b) No, 1.8462
c) Yes, 9.2308
d) No, 9.2308

29. In a linear regression equation with one independent variable, the p-value comes out to be
0.026. Which of the following statements is/are accurate?
I. The null hypothesis will be rejected at 5% level of significance for a two-tailed test
II. At 5% level of significance, we fail to reject the null hypothesis for a two-tailed test
III. The null hypothesis will be rejected at 1% level of significance for a one-tailed test
IV. The null hypothesis will be rejected at 10% level of significance irrespective of the tail of
the test
a) I and III
b) II and III
c) I and IV
d) II and IV

©2012 Konvexity All Rights Reserved Page 6

30. Which of the following statements is/are least accurate for the confidence interval for the
predicted value in a linear regression equation with one independent variable?
I. The confidence interval expands with an increase in the number of observations in the
sample
II. The confidence interval shrinks with an increase in the number of observations in the
sample
III. The confidence interval expands with an increase in the variance of residuals
IV. The confidence interval shrinks with an increase in the variance of residuals
a) I and III
b) I and IV
c) II and III
d) II and IV

31. Which of the following statements is/are least accurate about the conditional
heteroskedasticity?
I. The residuals increase with increase in value of independent variable
II. The residuals decrease with increase in value of independent variable
III. The coefficient estimates are not affected
IV. There are too many Type II errors
V. There are too many Type I errors
a) II and IV
b) II and V
c) Only IV
d) II, III and V

32. Which of the following statements is most accurate regarding the students' t-distribution test
for a linear regression equation with one independent variable? Assume that the sample mean
and sample variance remain the same on increasing the number of observations.
a) For a given level of significance, with an increase in number of observations, the chances of
rejection of null hypothesis increases because of both increase in degrees of freedom and
decrease in standard error
b) For a given level of significance, with an increase in number of observations, the chances of
rejection of null hypothesis decreases because of both increase in degrees of freedom and
decrease in standard error
c) For a given level of significance, with an increase in number of observations, the chances of
rejection of null hypothesis increases because of increase in degrees of freedom and the
chances of rejection of null hypothesis decreases because of decrease in standard error
d) For a given level of significance, with an increase in number of observations, the chances of
rejection of null hypothesis decreases because of increase in degrees of freedom and the
chances of rejection of null hypothesis increases because of decrease in standard error

©2012 Konvexity All Rights Reserved Page 7

33. The coefficient of determination for a multiple regression equation with 3 independent variables
is 0.7428. The total sum of squares, TSS, is 472. What is the standard error of the regression for
a sample size of 24?
a) 2.4637
b) 4.1869
c) 6.3613
d) 10.8105

34. Which of the following statements is least accurate about the adjusted R2?
a) The adjusted R2 in question no. 33 is 0.7048
b) The adjusted R2 approaches R2 on increasing the number of observations
c) The adjusted R2 approaches R2 on increasing the number of independent variables
d) The adjusted R2 may increase or decrease on decreasing the number of independent
variables

35. Which of the following statements is/are least accurate about multicollinearity?
I. Too many Type I errors
II. Too many Type II errors
III. Low correlation among the independent variables suggest the absence of
multicollinearity
IV. There is potential of multicollinearity if the t-tests indicate that the individual
coefficients are significantly different than zero while the coefficient of determination is low
a) I and III
b) II and III
c) I, III and IV
d) II, III and IV

36. Which of the following is/are least likely to be assumption of multiple regression?
I. The variance of error terms is zero
II. The independent variables are random
III. The expected value of error term is constant
IV. The error terms are not correlated with each other and are normally distributed
a) I and II
b) II and III
c) I and III
d) I, II and III

37. For a multi-factor model, Jesse Pinkman wants to test the null hypothesis that the intercept
term is greater than or equal to 4% using 5% level of significance. The number of factors used in
the model are 5 and the sample size is 35. In the regression result, the intercept term comes out
to be 8% with a standard error of 1.5%. What is the test-statistic value and the result of the
hypothesis?
a) 1.699 and fail to reject the null hypothesis
b) 1.699 and reject the null hypothesis
c) 2.667 and fail to reject the null hypothesis
d) 2.667 and reject the null hypothesis

©2012 Konvexity All Rights Reserved Page 8

38. The confidence interval for a regression coefficient in a multiple regression equation is -0.12 to
0.36 at 5% level of significance. What can we say about the result of the hypothesis test for
testing that the regression coefficient is significantly different than zero at 1% level of
significance?
a) Reject the null hypothesis
b) Fail to reject the null hypothesis
c) Reject the null hypothesis at 5% level of significance and can't say at 1% level of significance
d) Fail to reject the null hypothesis at 5% level of significance and can't say at 1% level of
significance

39. What is the value of F-statistics for a multiple regression having 3 independent variables? The
total sum of squares is 520 and the standard error of regression is 4. The number of
observations is 28.
a) 1.81
b) 2.82
c) 2.83
d) 22.59

40. Gustavo Fring wants to check whether there is significant restriction or not. The unrestricted
coefficient of determination is 0.72 and the restricted coefficient of determination is 0.60. The
number of restriction is 3 and the number of observation is 32. The unrestricted number of
independent variables is 6. What is the homoskedasticity-only F-statistic value?
a) 1.19
b) 1.29
c) 3.57
d) 10.71

41. According to which property of estimator, the standard error approaches zero as the sample size
approaches infinity?
a) Unbiasedness
b) Consistency
c) Reliability
d) Efficiency

42. Which of the following situations is least likely to result in the misspecification of a multiple
regression model with annual returns as the dependent variable?
a) Failing to transform an independent variable while the transformed variable is linearly
related to the annual returns
b) Omitting an independent variable that is related to the annual returns
c) Making a single regression model when in actuality the economic scenarios have changed
during the period
d) Not using lagged dependent variable as an independent variable

©2012 Konvexity All Rights Reserved Page 9

43. Which of the following statements is most accurate?
a) The probability of Type I error increases with an increase in level of significance
b) The probability of Type I error decreases with an increase in level of significance
c) The probability of Type II error increases with an increase in level of significance
d) The probability of Type II error decreases with an increase in level of significance

44. The return on a stock follows a Geometric Brownian motion. The mean return on the stock is 2%
over a horizon of 200 days and a standard deviation of 20% for 200 days period. The stock is
currently trading at $24. What will be the price of the stock at the end of the first day using the
random generated number as 0.084?
a) $24.031
b) $24.128
c) $24.167
d) $26.016

45. Which of the following statements is least accurate for Cox, Ingersoll, Ross model?
a) It allows for mean reversion of interest rates to some base level
b) The variance is proportional to the level of interest rates
c) It is a one-factor model of interest rate dynamics driven by long-term interest rates
d) The variance declines as interest rates move toward zero

46. Which interest rate dynamics model corrects the possibility that the short term and long term
interests rates do not drift away unrealistically from each other over time?
a) Cox, Ingersoll, Ross model
b) Brennan and Schwartz model
c) Co-integration model
d) Both a and b

47. The daily volatility of a stock return is estimated to be 1.2% and today's stock market return is -
12.0%. What is the new estimate of volatility using the EWMA model assuming a decay factor of
0.94?
a) 1.85%
b) 2.70%
c) 3.16%
d) 4.14%

48. What is the implied long-run volatility level of the returns of a stock following GARCH(1,1)
model? The model put a weight of 5% on previous period's return and 90% on the previous
volatility estimate. The weighted long-run variance is 0.000004.
a) 0.800%
b) 0.894%
c) 0.962%
d) 1.002%

©2012 Konvexity All Rights Reserved Page 10

49. Which of the following statements is least accurate about the persistence factor for a
GARCH(1,1) model?
a) The persistence factor should be lesser than 1 for the model to be stationary
b) A persistence factor of 1 means that there is no reversion and with each change in volatility,
a new level is attained
c) A model having a persistence factor of 0.94 takes longer time to revert back to the mean
value than another mode having a persistence factor of 0.95
d) The persistence describes the rate at which the volatility will revert to its long-term value
following a large movement

50. The EWMA is a special case of a GARCH(1,1) model with

a) ω = 0, α = 1 - λ, and β = λ
b) ω = 0, α = λ, and β = 1 - λ
c) ω = 1, α = 1 - λ, and β = λ
d) ω = 1, α = λ, and β = 1 - λ

©2012 Konvexity All Rights Reserved Page 11