Stock Market Integration: Granger Causality Testing with Respect to Nonsynchronous Trading Effects

Abstract

In this paper, we perform Granger causality analysis on stock market indices from several Asian, European, and U.S. markets. Using daily data, we point out the potential problems caused by the presence of nonsynchronous trading effects. We deal with two kinds of nonsynchronicity – one induced by differing numbers of observations in the series being analyzed and the other related to the different time zones in which the markets operate. To address the first problem, we propose a data-matching process. To address the second problem, we modify the regressions used in the Granger causality testing. When comparing the empirical results obtained using the standard technique and our modified methodology, we find substantially different results. Most of the relationships that are subject to nonsynchronous trading are not significant in the general case. However, when we use the adjusted methodology, the null hypothesis of a Granger non-causal relationship is rejected in all cases.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 43 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.