Academic journal article Economic Inquiry

The International Price Transmission in Stock Index Futures Markets

Academic journal article Economic Inquiry

The International Price Transmission in Stock Index Futures Markets

Article excerpt

I. INTRODUCTION

Numerous studies have investigated market linkages and price transmission mechanisms in major international equity markets, employing the analytical framework of the vector autoregression (VAR) or its variant, the error-correction model (ECM). (1) Studies such as Von Furstenberg and Jeon (1989), Eun and Shim (1989), and Koch and Koch (1991) focus on the short-run dynamic pattern of price transmission; others like Taylor and Tonks (1989) and Francis and Leachman (1998) are primarily interested in the long-run pattern of price transmission. More recently, an increasing number of studies explore both long- and short-run patterns of price transmission. Included in this last set are the works of Malliaris and Urrutia (1992), Arshanapalli and Doukas (1993), Masih and Masih (2001), and Bessler and Yang (2003), among others.

This study extends the examination of international price transmission to stock index futures markets. The article contributes to the existing literature in three aspects. First, a relatively new empirical framework is applied to allow for inferences of price transmission at three different time horizons: instantaneous, the short run, and the long run. Building on recent advances in statistical analysis of causal modeling using directed acyclic graphs (DAGs) as in Spirtes et al. (2000), Pearl (1995, 2000), and Swanson and Granger (1997), this study is able to explore the contemporaneous causal pattern underlying the correlations among market innovations. The existence of strong contemporaneous correlations among market innovations has been well documented in the United States and international stock markets by Agmon (1972), Eun and Shim (1989), Koch and Koch (1991), Housbrouk (1995), and Bessler and Yang (2003). It is also well recognized by Agmon (1972, 849) and Eun and Shim (1989, 246) that contemporaneous correlations among market innovations reflect the phenomenon that new information in one market is transmitted and shared by other markets in contemporaneous time, due to immediate response to price changes between markets. However, more in-depth analysis on exactly how instantaneous price transmission among market innovations is conducted in international equity markets has not yet been well addressed in the existing literature. Although Bessler and Yang (2003) touch on the issue, the necessity of imposing constraints in the spirit of the block-recursive structure noted by Koch and Koch (1991) in the DAG analysis of VAR innovations is proposed and discussed thoroughly in this study.

Second, innovation accounting analysis is more thoroughly explored in the study. Innovation accounting tools (i.e., impulse response analysis and forecast error variance decomposition) have been commonly used to summarize the dynamic pattern of price transmission among international financial markets. The importance of the factorization of innovations (i.e., VAR residuals) in yielding sound inference has been well acknowledged theoretically by Bernanke (1986), Sims (1986), and Swanson and Granger (1997). The application of the DAG technique, as discussed in Swanson and Granger (1997) and explained in the next section, is further key to innovation accounting analysis. In this study, the instantaneous price transmission pattern between market innovations (as identified by the DAG analysis) provides a data-determined solution to the basic problem of orthogonalization of residuals from the ECM and thus is critical to impulse response analysis or forecast error variance decompositions. Swanson and Granger (1997) argue that compared to the Choleski decomposition, the DAG-based structural decomposition is sensible but not subjective, because it allows for the properties exhibited by the data. Although several recent studies, such as those by Bessler and Yang (2003), Bessler et al. (2003), Haigh and Bessler (forthcoming), and Yang (2003), have used the DAG-based structural decomposition in a similar setting, the study is the first attempt responding to the suggestion by Swanson and Granger (1997, 364) of investigating the empirical implications of the DAG-based contemporaneous causal modeling. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.