Acceso usuarios
Lucie Samson
In this article, the restrictions imposed on excess returns by a latent variable model and an observed variable model are tested on stock market data from Canada and the United States. These two economies are highly integrated at the trade and production levels and it is to be expected that this is reflected in returns determination. The proposed latent variable model implies that all excess returns should move proportionately if assets are perfectly integrated. In our empirical analysis, data is disaggregated into ten size portfolios for each country. The restriction that all portfolios are governed by one single latent variable is rejected over the 1962-2004 sample period. It is established that this rejection is due to the presence of the smaller size portfolios. However, it is observed that Canada-US stock market integration has increased for small firms in more recent years. Financial and economic contributing risk factors are also identified.
