Dec

18

The attached chart shows this year's (1/5/10-12/16/10) compounded growth of SPY: "day" = o-c, "night" = c-o.

Both equity curves have converged toward the end of the year, but at various points there were divergences. From April to July, cpd day returns dropped from +6% to -9%, while overnights ended relatively flat. From July to present, day has risen steadily - now around +4%. Overnights dropped over the month of August, from +3% to -4 % - in the lead up to the QE2 announcement in late August.

It's hard to guess who figures into difference between intra-day and overnight returns, but to the extent that YTD returns are similar, the much smaller peak/valley of overnight suggests they were more accurate.

Two-sample T for c-o vs o-c

      N     Mean    StDev  SE Mean
c-o  241  0.00026  0.00702  0.00045 T=0.08
o-c  241  0.00020  0.00883  0.00057

>> statistically not different, but different volatility:

Test for Equal Variances: c-o, o-c

95% Bonferroni confidence intervals for standard deviations

      N      Lower      StDev      Upper
c-o  241  0.0063668  0.0070199  0.0078160
o-c  241  0.0080116  0.0088334  0.0098353

F-Test (normal distribution)
Test statistic = 0.63, p-value = 0.000

Levene's Test (any continuous distribution)
Test statistic = 10.10, p-value = 0.002

  


Comments

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 = '5654' AND comment_approved = '1' ORDER BY comment_date

Name

Email

Website

Speak your mind

Archives

Resources & Links

Search