Appropriate lag specification for daily responses of international stock markets

Abstract

This paper explores the international linkage of stock prices, using daily stock price indices of the four major economies (USA, UK, Germany, and Japan) from June 1974 to December 1997. It is argued that previous studies have not estimated the structural equation system reflecting the sequential occurrence of market closing, which is crucial in investigating the characteristics of daily responses among international stock markets. By estimating the structural equation system, it is found that the most recent market has the strongest effect, except for the case of Japanese effects on the German market.

Estimation of the common and country-specific shock to stock prices

“…First, when one estimates a VAR model using daily data, one is likely to obtain statistically significant coefficients on past returns (e.g. Tsutsui and Hirayama, 2003b). Although this may point to violation of market efficiency, the peculiarity of daily observations justifies this result.…”

Cointegration relationship and time varying co-movements among Indian and Asian developed stock markets

“…We may note further the time difference among these countries, for instance Tokio time is 3.30 hours ahead the Mumbai (that is the Indian city where is located the BSE) time, while both Singapore and Hong Kong are 2.30 hours ahead Mumbai time. Tsutsui and Hirayama (2004) examining the relationship among four major stock markets (Germany, Japan, UK and USA) point out that the specification of a lag structure of a VAR model must take into account both different closing prices and time differences among these countries in order to obtain better estimates of the integration among these markets. In order to check the robustness of our results, we also follow the study of Tsutsui and Hirayama (2004), assuming that the impact of time differences and closing prices is relevant also in our work.…”

“…Tsutsui and Hirayama (2004) examining the relationship among four major stock markets (Germany, Japan, UK and USA) point out that the specification of a lag structure of a VAR model must take into account both different closing prices and time differences among these countries in order to obtain better estimates of the integration among these markets. In order to check the robustness of our results, we also follow the study of Tsutsui and Hirayama (2004), assuming that the impact of time differences and closing prices is relevant also in our work. Schotman and Zalevzka (2006) point out that a way of dealing with this problem is to choose between daily data with potential time-matching problem and low frequency data (weekly or monthly)…”

International transmissions in US-Japanese stock markets

“…Though the analysed markets and methodology are different Ewing (2002) and Rigobon and Sack (2004) have a similar approach to our analysis here. Tsutsui and Hirayama (2004a) estimate a structural equations system using a seemingly unrelated system (SUR) of equations with the four major equity markets, Japan, Germany, the United Kingdom and the United States. Through the estimated structural equations system they find that there are many significant relationships in daily equity markets.…”