在實證研究上，本文應用的時間序列分析方法包括ADF，PP單根檢定、以及Johansen共積檢定、Granger Causality 因果關係檢定，最後並應用Copula相關檢定。
實證結果發現，(1).單根檢定：發現無論次級房貸前、期間均有單根現象，但取一階差分後就呈現穩定 (2). Johansen 共積檢定結果：發現二個市場指數在次級房貸前，期間均有共積現象，在次級房貸期間尤其明顯。(3)因果關係檢定：結果亦發現二市場具有因果關係 (4).最後透過copula-AR-GARCH檢定，發現二市場均有相關，但次級房貸期間，這現象更加明顯，意謂危機期間的影響效果變大。
The main purpose of this thesis is to analyze the influence, and, if possible, the degree of this influence, of the American stock market on the Canadian stock market. In consequence, the NASDAQ’s stock exchange and the Toronto stock index data were used.
To carry out this study, econometrical methods and tests were used. The tests thought to be the best fitted ones for this subject were the ADF, PP Unit Root test, the Johansen co-integration test, the Granger Causality test and, finally, the Copula correlation test.
First, the unit root test showed that, for both sub-samples, the data has unit root but when taking its first difference the data has no longer unit root and becomes stationary. Concerning the Johansen test, again, in both sub-samples, it was found that there is a co-integration between NASDAQ stock exchange and Toronto stock index. This result was verified when the Granger Causality was executed. Finally, the copula correlation test also showed that, after the U.S. subprime mortgage crisis, the correlation between the two stock markets was more important .This means that the co-movement was larger after the crisis hit the American economy.
Finally, we can conclude that the NASDAQ stock exchange has always had an influence over the Toronto stock index. Nevertheless, this influence became more apparent and gained strength after the U.S. subprime mortgage crisis hit the American economy.