Témata prací (Výběr práce)Témata prací (Výběr práce)(verze: 384)
Detail práce
   Přihlásit přes CAS
Influence of stock market variables on correlations among S&P sectors
Název práce v češtině: Vliv proměnných akciového trhu na korelace mezi sektory S&P
Název v anglickém jazyce: Influence of stock market variables on correlations among S&P sectors
Klíčová slova: korelace, DCC, S&P, výnos dluhopisu, ropa, VIX
Klíčová slova anglicky: correlation, DCC, S&P, bond yield, crude oil, VIX
Akademický rok vypsání: 2020/2021
Typ práce: diplomová práce
Jazyk práce: angličtina
Ústav: Institut ekonomických studií (23-IES)
Vedoucí / školitel: PhDr. František Čech, Ph.D.
Řešitel: skrytý - zadáno vedoucím/školitelem
Datum přihlášení: 25.05.2021
Datum zadání: 25.05.2021
Datum a čas obhajoby: 15.06.2022 09:00
Místo konání obhajoby: Opletalova - Opletalova 26, O105, Opletalova - místn. č. 105
Datum odevzdání elektronické podoby:01.05.2022
Datum proběhlé obhajoby: 15.06.2022
Oponenti: doc. PhDr. Jozef Baruník, Ph.D.
 
 
 
Kontrola URKUND:
Seznam odborné literatury
Ang, A. & Bekaert, G. (1999). International asset allocation with time-varying correlations, NBER Working Paper
7056.

Arouri, M. et al. (2011). Volatility spillovers between oil prices and stock sector returns: Implications for portfolio
management. Journal of International Money and Finance, 30, (7), 1387-1405.

Broadstock, D. C. & Filis, G. (2014). Oil price shocks and stock market returns: New evidence from the United
States and China. Journal of International Financial Markets, Institutions and Money, 33(C), 417-433.

Engle, R. F. & Sheppard, K. (2001). Theoretical and empirical properties of dynamic conditional correlation
multivariate garch. Working paper no. 8554, National Bureau of Economic Research.

Engle, R. F. (2000). Dynamic conditional correlation - a simple class of multivariate garch models. Discussion
Paper, 2000-09. University of California, San Diego, May 2000.

Engle, R. F. (2002). Dynamic conditional correlation. Journal of Business and Economic statistics, 20(3), 339–
350.

Gjika, D. & Horváth, R. (2013). Stock market comovements in Central Europe: Evidence from the asymmetric
DCC model. Economic Modelling, 33, 55–64.

Mensi, W. et al. (2016). Global financial crisis and spillover effects among the U.S. and BRICS stock markets.
International Review of Economics & Finance, 42, 257–276.

Syllignakis, M.N. & Kouretas, G.P. (2011). Dynamic correlation analysis of financial contagion: evidence from the
Central and Eastern European markets. International Review of Economics and Finance 20 (4), 717–732.

Wang, P. & Moore, T. (2008). Stock market integration for the transition economies: time-varying conditional
correlation approach. The Manchester School 76, 116–133.
Předběžná náplň práce v anglickém jazyce
Motivation:

Determining the correlations between financial instruments has been the cornerstone of portfolio creation.
Researchers over the past decades have focused on various ways of estimating the standard deviations
(volatility) as well as correlations among different financial assets. Vast majority of studies analyzing the
correlations investigated the co-movements of just aggregate stock indices from different countries. For example,
Syllignakis, M.N. & Kouretas, G.P. (2011) examined the interconnection between stock markets in Germany, the
US and Russia, and seven emerging countries from Central and Eastern Europe. Applying weekly data from
1997 to 2009 in the DCC-GARCH model, the authors show that returns on the US and German stock markets
significantly influenced the returns on the CEE markets. On the other hand, the correlation between the Russian
stock market and the CEE markets did not turn out to be statistically significant.

Next, Mensi, W. et al. (2016) examined the dynamic correlations between the stock markets in the US and
BRICS countries (Brazil, Russia, India, China and South Africa). Using the data between September 1997 and
October 2013, the results of the DCC-FIAPARCH model suggest that the US stock market returns were
significant for forecasting returns on stock markets in all BRICS countries. Moreover, the correlations between the
US and all BRICS countries except for Russia strengthened during and after the 2008-2009 financial crisis.

However, not much work has been devoted to the interrelation between individual sector or industry indices within
one country. The main goal of this thesis is to examine the time-varying dependencies among S&P 500 sectors.
All eleven sectors, represented by sector indices, will be considered in the analysis, whereas six sectors will be
studied in more detail. These six sectors are selected such that both cyclical and non-cyclical sectors are
represented. Further, although the precise distinction between value vs. growth sectors is not possible, some of
the chosen sectors tend to be tilted to value, or growth, respectively. Hence, the dynamics of dependencies
among cyclical, non-cyclical, rather-value, and rather-growth sectors will be observed and analyzed.

This research will aim to answer the question of how the analyzed sectors co-move when the overall market,
represented by the S&P 500 Index, goes up or down. The general notion is that stocks are correlated more in
times of market corrections or crises (Ang, A. & Bekaert, G. (1999)). However, this thesis is supposed to show
the difference in reactions of individual correlation pairs to the returns of the S&P 500 Index. Next, a possible link
between the US bond market (represented by the 10-year Treasury Note) and the dynamics of analyzed
correlations will be examined. Arouri, M. et al. (2011) as well as Broadstock, D. C. & Filis, G. (2014) showed that
sector stock indices might react to the oil market evolution differently. Hence, in my thesis I will also investigate
how oil price changes affect the correlations among examined S&P 500 sectors.


Methodology:

I will work with the daily data on eleven S&P 500 sector indices, the 10-year Treasury Note, and oil prices. The
data is available on the web page of S&P Global, The Wall Street Journal or Investing.com. To include two
considerable market downtrends – the ones in 2007-2009 and 2020 – time period between January 2007 and
February 2022 will be examined. Focusing on six sectors will lead to the detailed analysis of fifteen time-varying
correlations.

The ultimate selection of sectors is Consumer Staples, Energy, Financials, Health Care, Information Technology,
and Utilities. Consumer Staples, and Utilities represent non-cyclical sectors, whereas the remaining four sectors
are considered cyclical. Concerning the distinction between value and growth sectors, Financials (value) and
Information Technology (growth) are regarded as major counterparts. Next, Health Care is tilted toward the
growth factor, while the other three are rather value sectors. (S&P Global web page)

For each sector index, the logarithmic daily returns will be calculated. An appropriate form of ARMA-GARCH
model will be estimated, which will then become a basis for a multivariate GARCH model (e.g. Dynamic
Conditional Correlation introduced by Engle, R. F. (2000)). Time-varying correlations suggested by this model will
be further analyzed. First, the descriptive statistics of dependencies between individual sectors will offer a general
overview and comparison of the examined pairs.

Next, regression models will examine the relationship between the correlation pairs and the S&P 500 Index as
well as the 10-year Treasury Note or the oil price changes. These regression models will be applied for three
different lengths of returns: 1 day, 1 week, 1 month to see for which returns there are significant relationships
between the independent variable (S&P 500 Index, 10-year U.S. Treasury Note, oil price changes) and the
dependent variable (correlations among analyzed S&P 500 sectors).


Expected contribution:

This thesis should bring a deeper understanding of the dynamics of the interdependencies among S&P 500
sectors. Compared to studies that focused on correlations of aggregate stock indices, such as Gjika, D. &
Horváth, R. (2013) or Mensi, W. et al. (2016), my thesis will offer the sectoral perspective on the US stock
market. As both cyclical and non-cyclical sectors will be studied in more detail, this paper is supposed to explain
how correlations between these two types of sectors evolve. Furthermore, it should explore the co-movement
between sectors that are rather-value and rather-growth oriented.

Overall, the goal of this work is to deliver a better perception of the stock market – how the cyclical and noncyclical
sectors are related, in which situations value and growth indices are less (more) correlated, how the U.S
bond market or oil price changes influence individual correlation pairs, etc. Therefore, it should help investors with
construction of a diversified portfolio based on the specific phase, in which the stock market currently stands.

Moreover, the year 2020 was a year full of records and extremes as far as the stock market and events related to
it are concerned: oil price turned negative for the first time, the S&P 500 Index dropped by 34% between
February and March (23 trading days), then a 68% increase in the index until the end of the year followed, and
many other extremes. Maybe, this year has changed the style of investing. Therefore, it might be beneficial to
show the results of the analysis covering the years 2020 as well as 2021. Perhaps, the correlations in and after
2020 will considerably differ from their values 10 years ago, for example. That all shall be shown by this thesis.


Hypotheses:

1. Hypothesis #1: Correlations among S&P 500 sectors increase in times of a market decline.
2. Hypothesis #2: US bond market significantly influences the dynamics of the correlations among S&P 500
sectors.
3. Hypothesis #3: There exists a significant relationship between oil price changes and correlations among
analyzed S&P 500 sectors.

Outline:

1. Introduction: I will introduce the topic and show why it is important to do further research in this area.
2. Literature review: An overview of previous studies and their conclusions will be provided in this part.
3. Data: In this section, I will describe the data I will work with. Also, the procedure of sectors selection will
be discussed.
4. Methodology: This part will be devoted to the methodology I will follow in my research.
5. Results interpretation: In this section, I will present the results of my analysis, and show the differences
among the examined correlation pairs.
6. Conclusion: In this part, I will summarize the main results and goals achieved by this thesis.
 
Univerzita Karlova | Informační systém UK