This paper uncovers the relationship between stock markets and exchange rates in seven countries by employing stable aggregate currency (SAC) for the period of 1973- 2004. Ordinary Least Squares (OLS) regression, time series methods, and directed acyclic graphs are applied to the daily data on stock market indices and exchange rates. The findings based on regression analysis show that exchange rate exposure of stock markets is statistically significant when stock indexes in SAC are used. Using an innovation accounting technique, we confirm that stock markets and exchange rates are correlated. Moreover, in most cases stock markets are more exogenous than foreign currency markets, which explains the relatively high percentage of uncertainty in the foreign currency market. Overall, SAC-based models give relatively more accurate and robust results than those which employ stock indices in local currencies, because it is more accurate to convert both variables into the same denominator.
