Quantitative Methods for Finance exam
3 hrs 4 question, the exam time is after 6 hrs
1 of 6
Department of Accounting and Finance
M.Sc. Finance
M.Sc. Investment and Finance
M.Sc. International Banking and Finance
And
M.Sc. International Accounting and Finance
AG909: Quantitative Methods for Finance1
SAMPLE OF EXAM QUESTIONS
10.30am – 1.30pm (3 hours) + 25%
Instructions for Candidates – Please Read Carefully
Answer FOUR Questions
TWO from Section A and TWO from Section B
Section A counts for 40% of the exam marks. Section B counts for 60% of the exam marks
1. Financial calculators and scientific calculators with statistical functions are not allowed.
2. For quantitative (numerical) questions, present all the calculations and steps required to answer the questions.
1 This exam accounts for 70% of your final mark in this class
2 of 6
Section A
Answer TWO Questions
Q1. The Table below shows Government Spending (% of GDP) for Argentina and Spain for the years of
2005 until 2011.
a) Calculate the covariance and the correlation coefficient between Government Spending in the
two countries. Interpret both measures. Explain which of the two measures is better suited to
assess the strength of the association between two series.
(10 marks)
b) Test if Government Spending in Argentina and in Spain are equal. Clearly state the hypotheses
being tested, the conclusion of the test, and fully develop all calculations required (Use a 5%
two-tail Critical c value of the t distribution equal to 2.17).
(10 marks)
(TOTAL 20 MARKS)
Q2. Regarding the simple and multiple regression models:
a) Explain the meaning of a ceteris paribus effect and how it is linked with a partial effect. What is
the role of the zero conditional mean assumption in the interpretation of a partial effect?
(10 marks)
d) Explain the concept of multicollinearity. How can it be detected and prevented?
(10 marks)
(TOTAL 20 MARKS)
[Please Turn Over]
3 of 6
Q3. The regression output below
corresponds to the following equation estimated for a sample of returns for the IBOVESPA Index
(IBOV) and the S&P 500 Index (S&P500):
𝐼𝐵𝑂𝑉% = β0 + β1𝑆&𝑃500% + 𝑢%
a) Assuming a 1% significance level, calculate the difference in the expected return on the
IBOVESPA Index when the return on the S&P500 increases from 0.25. to 0.30.
(10 marks)
b) Calculate and interpret the regression’s R-square. Based on its explanatory power, how does the
model fare in terms of goodness of fit. Justify your answer.
(10 marks)
(TOTAL 20 MARKS)
[Please Turn Over]
4 of 6
Section B
Answer TWO Questions
Q4. The regression output below
corresponds to the following equation estimated for a sample of Korean firms:
Inv = β0 + β1 𝐶𝑎𝑠ℎ𝑓𝑙𝑜𝑤 + β2 𝐶𝑎𝑠ℎ 𝑓𝑙𝑜𝑤5 + u
where Inv (Investment), and Cash flow is the independent variable. The sample averages for
Investment and Cashflow are 𝐼𝑛𝑣=0.237 and 𝐶𝑎𝑠ℎ𝑓𝑙𝑜𝑤=0.026, respectively. Cash flow is included in
the model both in linear and quadratic forms. In the regression output table, the variable “Cashflow”
refers to the linear form, whereas the variable “Cashflow Squared” refers to the quadratic form.
a) Calculate the partial effect of Cashflows on Investment. Interpret the effect of Cashflows on
Investment. Are there increasing or decreasing marginal effects? Justify your answer.
(10 marks)
b) Test if β0 is statistically significant (Use a 5% two-tail Critical c value of the t distribution equal
to 1.960). Fully explain the hypotheses being test and develop all the steps involved in the test.
(10 marks)
c) Calculate the expected investment when Cash flows is at the sample’s average.
(10 marks)
(TOTAL 30 MARKS)
[Please Turn Over]
5 of 6
Q5. The regression output below
corresponds to the following equation estimated for a sample of 38 countries:
Inv/GDP = β0 + β1Credit/GDP + 𝛿0Emerging + β2Emerging x Credit/GDP + u
where Inv/GDP is the ratio of domestic investment to GDP (gross capital formation divided by gross
domestic product), Credit/GDP is the ratio of bank credit to GDP (bank credit to the private sector
divided by gross domestic product), Emerging is a dummy (qualitative) variable equal to 1 for emerging
economies, and equal to 0 for developed economies, and Emerging x Credit/GDP is an interaction
term between the dummy variable Emerging and Credit/GDP.
a) Assuming a 5% significance level, calculate the effect of Credit/GDP on Investment in Emerging
economies. Interpret the interaction effect.
(10 marks)
b) Assuming a 5% significance level, is the difference in Investment/GDP between developed
and emerging economies significant? In which group of countries Investment/GDP is higher?
(10 marks)
c) Structure a test of hypothesis to verify the statistical significance of the model. Show and
explain how the F statistic can be obtained. Assuming a 1% significance level, explain
whether the regression model is significant or not.
(10 marks)
(TOTAL 30 MARKS)
[Please Turn Over]
6 of 6
Q6. Regarding pooled and panel data models:
a) Briefly explain the importance of accounting for time effects in a Pooled model.
(10 marks)
b) Explain how can a fixed effects model tackle omitted variable bias. Can this model tackle both
unobserved and observed omitted variables?
(10 marks)
c) Assume a given explanatory variable is uncorrelated with the unobserved effect. May a
Random Effects model be suitable in this case? Justify your answer.
(10 marks)
(TOTAL 30 MARKS)
End of Paper