What’s really notable about these short term correlations is how they now peak at a much higher level than they did before 2015. This is likely an artifact of two things. One is a data availability issue -- I can only calculate correlations using current SPX constituents, which means that the sample size is somewhat limited when you go way back in time. The other is, of course, the rise of index or ETF-based investing, where a single trade can put selling pressure on hundreds of stocks and the same time.
What’s really interesting is that the one-week change in the 10-day correlation of SPX stocks has risen faster than ever before. This is admittedly a sort of cherry-picked time frame, but between those September 2 highs and Wednesday, September 9, the correlation rose from 0.9 to 0.79. The correlation has never risen that quickly in the history of my data set. In other words, never before have SPX stocks snapped into line this much, this fast. #stocktalk
What’s really interesting is that the one-week change in the 10-day correlation of SPX stocks has risen faster than ever before. This is admittedly a sort of cherry-picked time frame, but between those September 2 highs and Wednesday, September 9, the correlation rose from 0.9 to 0.79. The correlation has never risen that quickly in the history of my data set. In other words, never before have SPX stocks snapped into line this much, this fast. #stocktalk