Serial correlation coefficients, correlograms and periodograms are computed by taking cross products over fixed or variable time intervals. I believe a more salient way to do it for markets might be a variant of computing them with respect to price intervals.

On the same front, the empirical chance of the market's closing exactly unchanged, or very close, is much higher than would be expected from chance. And what does this indicate for the future? An idea comes from the expected utility of a vote in a election, its being very low considering the chances that you would cast the deciding vote.

The economy in England showed much vibrancy while I was there and all taxi drivers said that in the last two weeks there had been a marked improvement in activity from its previous relatively high levels. Perhaps taxi drivers should join truck drivers as opinion leaders to tally when looking for leading indicators of the economy's health.


WordPress database error: [Table './dailyspeculations_com_@002d_dailywordpress/wp_comments' is marked as crashed and last (automatic?) repair failed]
SELECT * FROM wp_comments WHERE comment_post_ID = '4009' AND comment_approved = '1' ORDER BY comment_date




Speak your mind


Resources & Links