Not So Important Days Any More? from Kora Reddy
Since Jan 1st 2000 to Dec 31st 2015
Total $SPY returns for all days 92.71 points
returns on 1dom + fed day (most important days) are 89.02 pts (about 97% returns captured by doing hard work on 322/4025 days (8%))
The year gone by 2016 is a different story.
Total $SPY returns are 23.95 pts and on most important days the returns are a mere 0.71 pts (20 of them out of 252 days)
So where did the returns shift to?
answer -> 11-15th trading day
2016 $SPY returns 23.95 pts
Returns on 11,12,13,14,15th trading days combined are 24.14 pts (60/252 = ~24 % days)
Conclusion, work load increased in 2016 from 8% to 24%, a 200 % rise in work hours. A 2017 wish is reduce in the lower working hours!
Kim Zussman writes:
I also checked whether the past 16 years were trendy for stocks.
If today's close was above 100D moving average, return from today's close to tomorrow's close. ">100DMA" (mean)
If today's close was below 100D moving average, return from today's close to tomorrow's close. "<100DMA" (mean):
Two-sample T for >100DMA vs <100DMA
N Mean StDev SE Mean
100DMA 2766 0.00025 0.00824 0.00016 T=-0.06
<100DMA 1394 0.0003 0.0179 0.00048almost the same. However, as expected daily returns were less volatile when the market was going up:
Test for Equal Variances: >100DMA, <100DMA
95% Bonferroni confidence intervals for standard deviations
N Lower StDev Upper
100DMA 2766 0.0079943 0.0082355 0.0084912
<100DMA 1394 0.0171828 0.0179134 0.0187065
F-Test (normal distribution)
Test statistic = 0.21, p-value = 0.000
Levene's Test (any continuous distribution)
Test statistic = 540.84, p-value = 0.000