There are three sections in this coursework assignment and you are required t
o
answer all questions.
Section A (30 marks)
Construct a portfolio of stocks from the US market and using EViews analyse its
performance based on the estimates from the Fama – F
rench (1992, 1993) and
Carhart (1997) models. You are required to answer the two questions listed below. Please follow the
instructions in these questions.
Requirements:
1) Select 10 stocks listed on the US market and explain your choice. You ma
y
apply such selection criteria as market indicators (P/E, P/B
V, DY ratios etc.) or
any other selection rule, which you can reasonably justify. The period of
analysis should be the last full 10 calendar years from January 2011 to
Draft Assessment Brief – Postgraduate Page 2 of 8
December 2020 and the data frequency should be monthly. Download the
stock price data from Bloomberg and construct a portfolio from your stocks.
Calculate the monthly returns of this portfolio in the entire 10-year period:
2011-2020.
Using Eviews perform estimation of parameters of Fama – F
rench and Carhar
t
models in the entire sample (10 years) co
vering 120 monthly obser
vations.
The data for the Fama – F
rench and Carhart models factors for the US market
can be found at Prof Kenneth F
rench’s website using the following link:
https://mba.tuck.dartmouth.edu/pages/faculty/ken.french/data_librar
y.html
The three Fama and F
rench (1992,1996) factors RmRf, SMB, HML can be
found in the .csv
file under the “Fama/F
rench 3 Factors” and the Carhart’
s
(1997) MOM factor can be found in the .csv
file under “Sorts Involving Prior
Returns” “Momentum Factor Mom”.
Divide the whole sample into 10 annual periods and investigate how the
estimation results change in these sub-samples in case of your portfolio.
Discuss the econometric issues, which are related to your estimations.
(15 marks, 300 words appr
ox.)
2) Discuss your estimation results in terms of the e
valuation of your portfolio
performance in the whole period and in all 10 annual sub-periods. Analyse the
existence of small stocks effect, value premium effect and momentum effect in your portfolio and explain your results in line with the asset pricing theories.
(15 marks, 700 words appr
ox.)
Draft Assessment Brief – Postgraduate Page 3 of 8
Section B (30 marks)
Using the data from the US stock market, analyse the extent of the inter
valling effect
in the estimation of stocks’ beta (
β) parameters.
You are required to answer the two questions listed below. Please follow the
instructions in these questions.
Requirements:
3) Select 15 companies from the US market and download their stock price data
from Bloomberg. The sample period should include the last full 5 calendar years: from 2016 to 2020. Your selection should cover companies of different
sizes, i.e. 5 large stocks, 5 medium size stocks and 5 small stocks. Discuss
brie
fly the stocks which you ha
ve chosen and then use the CAPM model (in
the market model version, i.e. without a deduction of the risk free rate) t
o
estimate in EViews their betas (
β) for daily and monthly frequency data.
Discuss the econometric issues, which are related to your estimations.
(20 marks, 200 words appr
ox.)
4) Summarize and discuss your
findings about the extent of the inter
valling effect
in the estimation of the beta (
β) parameter in case of all your 15 stocks and
discuss any patterns that you found across your results from the point of view
of the differences in daily and monthly beta (
β) estimates and the stocks size.
(10 marks, 800 words appr
ox.)
Section C (40 marks, 1000 words)
Using the data in the excel
file “Panel Data” uploaded on the modules Blackboar
d
site, identify and discuss the determinants of capital structure of the selected
firms.
The
file is uploaded under the “Assessment and Submission folder of the module’s
Blackboard Site”. The
file consists of following
firm-speci
fic factors: liquidity (nwc),risk (beta), dividend payment- pay dividend or not (dividend),
firm size(size), and
Draft Assessment Brief – Postgraduate Page 4 of 8
capital structure (lev). The “dividend” variable is a binary variable which takes a value
of 1 if the
firm pays dividend and 0 if the
firm does not pay the dividend. The data is
already in the panel data format.
Requirements:
1) Upload the data in Eviews and estimate the appropriate panel data model b
y
performing the required tests to identify the determinants of the capital
structure of
firms.
2) Clearly explain and discuss the model results