Essays on Financial Markets

Abstract

1. En el primer capítulo “Impact of oil prices on international financial markets” se analiza el impacto de cambios en el precio de petróleo (barril WTI) y su volatilidad en los principales mercados de renta variable: DJIA, S&P500, FTSE100, DAX y NIKKEI225. Se investiga la relación lineal y no-lineal. Se usan los precios diarios del periodo 1984 – 2005. Se observa que los cambios en el precio de petróleo afectan negativamente a los mercados americanos (DJIA, S&P500) y DAX, este impacto es negativo y representa un tercio del retorno diario. No se observa ni un efecto de retraso en esta relación ni de asimetría. Los mercados americanos reaccionan positivamente a la alta volatilidad de los precios del petróleo en el entorno de bajada del precio que se explica por menor factor de coste para las empresas. No se detecta relación entre los retornos de las bolsas y las transformaciones no-lineales del precio del petróleo. Se observa la transmisión de los shocks del precio del petróleo a la volatilidad de las bolsas analizadas.

2. En el segundo capítulo “Changes in correlations between CEE stock markets and European stock markets” se detecta el cambio estructural en la correlación entre las bolsas de los países de Europa Central y de Este (bolsas de Varsovia, Praga, Budapest, Bratislava y Liubliana) y las principales bolsas de Europa de Oeste (Frankfurt, Londres, Vienna) alrededor de Mayo 2004, unos meses antes de la entrada de estos países a la Union Europea. Para detectar este cambio estructural se usan los precios diarios de los índices y el modelo Asymmetric Dynamic Correlation GARCH que permite modelizar las correlaciones y utilizar las pruebas del cambio estructural. EL mayor nivel de las correlaciones entre las bolsas de los países de Este y Oeste, después de la entrada de los países europeos emergentes a la UE en 2004, es una clara señal de la mayor integración de los mercados financieros de los países de Este con los mercados europeos.

3. El tercer capítulo “Structural changes in the volatility of IBEX35” investiga los cambios estructurales en la volatilidad del índice principal español. Se considera el periodo 1992 – 201, el modelo asimétrico GARCH y el algoritmo ICSS, y se detectan varios periodos de alta y baja volatilidad que corresponden a los principales acontecimientos económicos de la economía española y mundial. Estos cambios de volatilidad se detectan también para otro mercado bursátil de referencia – DJIA. Se observa que saber en qué estado de la volatilidad se encuentra el índice bursátil mejora significativamente la calidad de las predicciones de la volatilidad, una variable muy importante a lo hora de valorar derivados.

Chapter 1, Impact of oil prices on international financial markets, analyzes the relation between oil price returns, volatility of oil price return and returns of stock indices. We consider daily prices of WTI crude and daily quotations of five main world stock indices - DJIA, S&P500, FTSE100, DAX and NIKKEI225. We investigate both the linear and non-linear relationship between oil price returns and stock market returns. We analyze also the links between the volatilities of the returns of oil prices and stock market returns. The changes in oil prices negatively affect DJIA, S&P500 and DAX. The increase in oil prices (positive returns) lowers the return on the stock index. Lagged oil returns have no influence on the stock market; neither there is any asymmetry in this relationship. The returns of DJIA and S&P 500 react to high volatility and falling oil prices. DAX reacts positively to low volatility of oil returns and negatively to high volatility of oil returns combined with increasing prices - the consequences of geopolitical events. In the same way NIKKEI also negatively reacts to the high oil return volatility. There is no relationship between the non-linear transformations of oil prices and the stock market returns. The positive shocks to oil price returns (thus increases in prices) are transmitted immediately to stock markets in form of negative shocks to stock market returns. There is also a feedback reaction of stock index volatility to the level of volatility of oil prices - the low level of oil volatility diminishes the volatility of stock market indexes next day.

Chapter 2, Changes in correlations between CEE stock markets and European stock markets, analyses the changes in correlations among Western European and CEE stock markets before and after the EU entrance in 2004. This analysis allows investigating the benefits from EU integration. In our analysis we consider the daily prices over the period 1994-2006 for the following stock market indexes - WIG20 (Poland), PX50 (Czech Republic), BUX (Hungary), SAX (Slovakia) and SBI (Slovenia), DAX30 (Frankfurt), FTSE100 (London) and WBI (Vienna). Using the Asymmetric Dynamic Correlation GARCH Model we estimate conditional correlation series between CEE stock markets and three important European stock indexes. Then we detect a structural break in the conditional correlation series some months before May 2004 - the date of the entrance of all the countries to the EU. The successive increase in correlation of the Central and Eastern European financial markets and Western Europe is the sign of the higher integration of these markets with European stock markets.

Chapter 3, Structural changes in the volatility of IBEX35, analyzes the changes in the volatility of the Spanish index IBEX 35 over all its history from 1992-2011. The changes in volatility levels of stock market indexes are important in many aspects of business and financial life. Using Quadratic GARCH model of Sentana (1995) and modified ICSS algorithm to detect the structural changes in the volatility of the index, we detect several structural breaks in the volatility of IBEX35. The subsamples defined by the breaks differ in the persistence and the asymmetry of the impact of shocks on volatility. We observe a better forecasting performance of the model with breaks than the benchmark, which is the QGARCH model estimated over the entire period of interest. We also observe the similar behavior of another important stock index – DJIA (Dow Jones Industrial Average).