Mitchell Ratner

Turn-of-month and pre-holiday effects on stock returns: Some international evidence

This study examines turn-of-month and pre-holiday effects on international markets. Turn-of-month effects are significant in Canada, the UK, Australia, Switzerland, and West Germany. Pre-holiday effects are significant in Canada, Japan, Hong Kong, and Australia. The absence of these effects in certain markets suggests that they originate from country-specific institutional practices. All countries exhibiting pre-holiday effects do so before local holidays; only Hong Kong does so before US holidays. This reinforces the conclusion that such anomalies are not generated solely by American institutions.

The Portfolio Implications of Gold Investment

This article investigates the use of gold as an investment asset. The data consist of U.S. and foreign equity returns from 1975 to 2005. The results indicate that investment in gold is inferior to a simple buy-and-hold strategy of U.S. equities over the long term. Gold is often believed to provide potential as a defensive asset, given its low correlation with U.S. equities. However, a portfolio optimization technique using actual and simulated data indicates that the long-term portfolio benefits of holding gold are marginal at best

A cointegration test of the impact of foreign exchange rates on U.S. stock market prices

Since the removal of the governmentally enforced fixed exchange rate standard in the early 197Os, exchange rates have been determined by the forces of the global marketplace. A strong relationship between the dollar and U.S. stock prices is a popular view in the financial news media. The Wall Street Joournal often associates a U.S. stock market rally with a change in the value of the dollar. Unexpected changes in exchange rates create uncertainty regarding a firm’s assets, liabilities and future cash flows. Multinational firms, firms that rely on foreign direct investment, and firms that conduct import or export trade are especially susceptible to exchange rate risks. Indeed, even purely domestic firms are potentially impacted through foreign competition. A reduction in the value of U.S. currency is expected to favorably affect U.S. stock prices due to increased exports and domestic substitution foi imported goods. Conversely, a strengthening dollar is expected to result in declining stock prices. The belief that a depreciating dollar will result in rising exports is based on the assumption that as the dollar begins to fall, it will continue moving in that direction for some time. The market will respond by reassessing expectations regarding future changes in exchange rates. Although this conviction is common among “technical analysts,” it is inconsistent with modern rational expectations theory As exchange rates are financial prices, they should only be affected by changes in expectations about economic activity In an efficient market, they already reflect all information about expected future economic activity and policy The purpose of this paper is to test if U.S. dollar exchange rates affect U.S. stock prices using cointegration analysis. Evidence is presented that demonstrates no systematic relationship between U.S. currency value and stock prices. The remainder

Investigating the behavior and characteristics of the Madrid Stock Exchange

Abstract This study examines the efficiency and characteristics of the nine major indexes of the Madrid Stock Exchange from 1941 through 1992. The findings do not support weak form efficiency among the General, Bank, Construction, Investment, Metal-working, and Chemical Indexes. Results of the Utility, Food, and Communications Index tests are not consistent.

Sector Dispersion and Stock Market Predictability

This paper investigates the lead/lag relationship between the variation of the 10 primary sector indexes (sector dispersion) with market returns and market volatility. The sample consists of U.S. data from January 1974 through December 2003. This study documents a statistically significant lead/lag relationship between sector dispersion and both market returns and market volatility. Asymmetry analysis reveals that high sector dispersion is a consistent predictor of market volatility. Dispersion is found to be an effective predictor of bear market volatility, and both bull market and bear market returns.

Cross-Autocorrelation in the Brazilian Equity Market Before and After Financial Liberalization

ABSTRACT This study examines cross-autocorrelation in the Brazilian equity market before and after the effect of its financial liberalization. The sample consists of daily data from January 1986 through December 1999. Differences in size-based portfolios are tested using Granger-causality and trading volume. The results confirm that large stock portfolio returns lead small stock portfolio returns, but the lagged response of small stock portfolio returns diminish following financial liberalization. Relative trading volume of small stocks increases after financial liberalization while relative trading volume of large stocks decreases. RESUMEN. Este estudio examina la auto correlación cruzada antes y después de la liberalización del mercado. La muestra consiste de datos diarios entre enero de 1986 y diciembre de 1999. Las diferencias entre las carteras formadas según el tamaño de las empresas se prueban usando la causalidad de Granger y el volumen negociado. Los resultados confirman que las carteras de acciones de empresas de gran envergadura anticipan el retorno de las carteras pertenecientes a empresas de pequeño porte, pero la respuesta desfasada de la cartera de empresas de pequeño porte disminuye después de la liberalización financiera. El volumen negociado inherente a las acciones de las empresas pequeñas aumenta después de la liberalización financiera mientras que el volumen negociado relativo a las acciones de las empresas de gran envergadura disminuye. RESUMO. Este estudo examina a auto correlação cruzada antes e depois da liberalização do mercado. A amostra consiste de dados diários entre janeiro de 1986 e dezembro de 1999. As diferenças entre carteiras formadas segundo o tamanho das empresas são testadas usando-se a causalidade de Granger e o volume negociado. Os resultados confirmam que carteiras de ações de empresas de grande porte antecipam o retorno de carteiras de empresas de pequeno porte mas a resposta defasada da carteira de empresas de pequeno porte diminui depois da liberalização financeira. O volume negociado relativo das ações de empresas pequenas aumenta depois da liberalização financeira enquanto o volume negociado relativo das ações de empresas de grande porte diminui.

Co-Movements of U.S. and Latin American Equity Markets in Bull and Bear Markets

ABSTRACT In this paper, we study the co-movements of the U.S., Argentine, Brazilian, Chilean, and Mexican equity markets during the October 12, 1998-March 24, 2000 bull market and the March 24, 2000-September 10, 2001 bear market. We find that these five markets are more closely correlated and, therefore, there is less diversification benefit to global investors during the bear market than during the bull market. Our findings, using rolling correlation analysis, indicate that the most volatile U.S. correlation coefficient is with Chile, both in the bull market and in the bear market. The U.S. correlation coefficient with Mexico is more stable (i.e., more predictable) during the bull market than during the bear market. However, the U.S. correlation coefficients with Argentina and Chile are less stable (i.e., less predictable) during the bull market than during the bear market. The degree of stability (i.e., predictability) of the U.S.-Brazilian correlation coefficient is about the same, both in the bull market and in the bear market. We use impulse simulation analysis to study the responses of the four Latin American markets to a simulated shock in the U.S. market. The shock causes strong responses in all four Latin markets. The Argentine and Mexican markets show a stronger response in the bull market than in the bear market. However, the Brazilian and Chilean markets show a stronger response in the bear market than in the bull market. The shock causes considerable turbulence in all markets. The turbulence lasts longer in the bear market than in the bull market. RESUMEN. Este documento estudia los co-movimientos ocurridos en los mercados bursátiles de Estados Unidos, Argentina, Brasil, Chile y México durante el período bajista ocurrido entre octubre 12 de 1998 a marzo 24 de 2000, y el mercado alcista entre marzo 24 de 2000 a septiembre 10 de 2001. Hemos encontrado que estos cinco mercados se encuentran estrechamente vinculados y, consecuentemente, ofrecen menores ventajas diversificadas para los inversores globales durante el mercado bajista que durante el período alcista. Para elaborar nuestro estudio utilizamos un procedimiento analítico de correlación renovable, que indicó que los mayores coeficientes de correlación se establecen con Chile, tanto en el mercado alcista como bajista. El coeficiente de correlación norteamericano con México es más estable (Ej: mayor grado de previsibilidad) durante el mercado alcista que el bajista. No obstante, los coeficientes de correlación entre Norteamérica y Argentina y Chile son menos estables (Ej: menor previsibilidad) durante el mercado alcista que durante el bajista. El grado de estabilidad (Ej: previsibilidad) de la correlación Estados Unidos-Brasil es aproximadamente el mismo tanto en el mercado alcista como bajista. Hemos utilizado un análisis de simulación de impulso para estudiar las respuestas dadas por los cuatro mercados Latinoamericanos, para simular el choque en el mercado norteamericano. El choque provoca fuertes respuestas en los cuatro mercados de América Latina. Los mercados de Argentina y México muestran una respuesta mucho más fuerte en el mercado en alza que cuando está en baja. Sin embargo, los mercados brasileño y chileno muestran una respuesta mayor en el mercado en baja que cuando está en alza. El choque causa una turbulencia considerable en todos los mercados, que dura por un período más prolongada en el mercado en baja que en el mercado en alza. RESUMO. Neste trabalho, estudamos os movimentos conjuntos dos mercados de ações norte-americano, argentino, brasileiro, chileno e mexicano, durante a alta do mercado, no período de 12/10/1998 a 24/03/ 2000, e a baixa do mercado, no período de 24/03/2000 a 10/09/2000. Percebemos que estes cinco mercados estão intimamente relacionados e, por isso, há menos diversificação dos benefícios para os investidores internacionais durante a baixa do mercado do que durante a sua alta. Nossa descoberta, através da análise da correlação flutuante, indica que o coeficiente de correlação americano mais volátil é em relação ao Chile, tanto na alta quanto na baixa do mercado. O coeficiente de correlação com o México é mais estável (i.e., mais previsível) na alta do que na baixa do mercado. Contudo, tais coeficientes, em relação à Argentina e ao Chile, são menos estáveis (i.e., menos previsíveis) nas altas e baixas do mercado. O grau de estabilidade (i.e., de previsibilidade) do coeficiente de correlação americano e brasileiro é semelhante, em ambos os mercados. Utilizamos a análise de simulação de impulso para estudar as respostas dos quatro mercados da América Latina a um choque simulado no mercado americano. O choque causa respostas fortes nos quatro mercados latinos. Os mercados da Argentina e do México apresentam reações mais fortes no mercado em alta do que no mercado em baixa. Contudo, os mercados do Brasil e do Chile já apresentam uma reação mais forte no mercado em baixa do que no mercado em alta. O choque causa uma turbulência considerável em todos os mercados. A turbulência perdura mais na baixa do mercado do que em sua alta.

Reducing Stock Risk with Hedge Funds

This paper tests the risk reduction properties of hedge fund investing against a sample of stocks ranging from 1990 through 2014. GARCH dynamic conditional correlation analysis indicates that hedge funds are a significant diversifier due to the consistent imperfect relationship between the hedge fund returns and the stock return indexes. Hedge funds serve as a weak safe haven in times of extreme stock market volatility. During periods of financial crisis, hedge funds also largely function as a weak safe haven. In contrast to their name, hedge funds do not provide a traditional “hedge” against stock risk.

Hedging Emerging Market Stock Risk with Sovereign Credit Default Swaps

This paper tests sovereign credit default swaps (CDS) as a hedge and safe haven asset against stock index returns in a sample of 18 emerging markets from 2005-2014. GARCH dynamic conditional correlation analysis indicates that CDS are an effective hedge due to the consistent inverse relationship between CDS and the stock indexes of each country. CDS are largely a weak safe haven in times of extreme stock market volatility. However, during the 2008 U.S. financial crisis and the 2010 European debt crisis, CDS provide either a strong or weak safe haven in most countries.

Gold and Systemic Risk

This study tests gold as a hedge and safe haven asset against systemic risk in 21 emerging and developed countries from 1979 to 2012. Generalized autoregressive conditional heteroskedasticity (GARCH) dynamic conditional correlation analysis indicates that gold serves as an effective hedge against systemic risk. Gold also provides a safe haven in times of extreme market volatility and during periods of financial crises in most countries.