A Critical Appraisal of Studies Analyzing Co-movement of International Stock Markets

Literature is reviewed on the analysis of co-movement between the price indices of stocks or their realized returns at various markets. Four major categories of frequently recurring methodological shortcomings are registered. These are: (i) omitted regressor problems, (ii) neglecting to verify agreement of estimation outcomes with adopted model assumptions, (iii) employing particular statistical tests in inappropriate situations and, occasionally, (iv) lack of identification. The devastating effects of the detected methodological defects are explained in mildly technical appendices and are also illustrated by simulations and empirical examples.