macrocephalopod

Technically you can get loads more data with overlapping intervals. Adjusting hypothesis tests for the autocorrelation structure is a bit of a mess. (not impossible though). Most people would never do this and conclude things entireyl wrongly.

If you only have the overlapping intervals that’s true for some use cases. But if you have the sub-periods too then the overlapping intervals can’t buy you any more power (for obvious info theory reasons).

Always fun to see “analysis” that looks at rolling monthly yoy instead of the monthly change and gets an r^2 about 3x overstated as a result

1/2 Hmm I get that intuitively the overlap means that not all samples are independent so true variance should be higher than sample variance (and thus t-stat lower), but it also intuitively feels like completely disjoint subperiods throws away info.

2/2 For vol estimation, the Hodges and Tompkins correction can be used to correct this bias (the bias in the variance of vol). Do you know if there's a similar method for general regressions?

newey west has entered the chat

Wait, why Newey West?