I wrote this article last year which has now become topical again because oil is once again making new all time highs. If you have any comments or questions then please reply to this post.

I ran across my statistics professor last night at school and showed her these charts. For what it's worth, she suggested lagging the S&P data to see if there is any correlation as an after-effect.

That's not a bad idea. I don't have time right now to do that but one could offset one of the data series by 1 time unit (1 day in this case) and recalculate the correlation coefficient to see if we get a higher or lower number (and therefore stronger correlation) and keep doing that to see which offset, both positive and negative, gives us the best correlation.

One of the problems with this is my observation that the correlation only appears when Oil is at an extreme (not sure if this observation was in the original article but may have been mentioned on the forum somewhere). So when Oil is not at an extreme the correlation is not as strong.

I'm not sure how you take a broken series like that and measure correlation when one of the series is approaching its all time high and making new highs. If you run into your statistics professor again then ask her that.

Will do.