The S&P 500
Interpretation is the revenge of the intellect upon art.
The calculation of the correlation coefficient between the index to be predicted and other related markets is the first step of a complete and thorough intermarket analysis. Of course, the final challenge comes from interpretation. Intermarket analysis relies on the premise that relationships in the past will be the same in the future so before making any conclusions it is always prudent to assess the stability of the correlation coefficient over time.
I chose to start this correlation analysis with a broad market index like the S&P 500 which is widely regarded as the best single gauge of the US equities market.
5.1 CORRELATION WITH INTERNATIONAL INDICES
The first step in a correlation or regression analysis is to decide whether you want to analyze the price on a daily, weekly, monthly or quarterly basis. Using long time scales has the added apparent theoretical benefit that averaging over long periods suppresses windowing errors and daily noise, revealing the underlying long term correlations. The choice of time periods is therefore very important and should depend on a trader’s time horizon. This is illustrated in Tables 5.3
where the correlation rises dramatically when weekly returns replace daily data. As the time segment is reduced, noise becomes a bigger factor in the calculation and the correlation drops due to an abundance of change which is unfolding in the short term.