Daily Seasonalities In S&P 500

Are there any daily seasonalities in S&P 500? I did some research into SPY, an ETF based on S&P 500 components and a very good proxy, based average return on which day of the month.

To get a sufficient number of data I started back in 2005. Here are the results:

 Average gain  Average gain Average gain Average 5 days Average 5 days Average 5 days
close to close open to close to Close Open to
close to open close close to open close
1 0.097 0.010 0.083
2 0.062 0.074 -0.014
3 0.134 0.099 0.037
4 -0.096 -0.044 -0.055
5 0.107 -0.088 0.195 0.060 0.010 0.050
6 -0.068 -0.069 -0.001
7 -0.271 0.034 -0.302
8 0.115 0.021 0.092
9 -0.136 0.099 -0.234
10 0.052 0.033 0.018 -0.060 0.020 -0.090
11 -0.034 -0.088 0.054
12 0.014 -0.140 0.157
13 0.426 0.170 0.248
14 -0.068 0.071 -0.137
15 -0.196 -0.095 -0.105 0.030 -0.020 0.040
16 0.233 0.123 0.110
17 -0.203 -0.046 -0.158
18 0.281 0.143 0.138
19 -0.149 0.122 -0.269
20 -0.254 -0.042 -0.216 -0.020 0.060 -0.080
21 0.127 0.161 -0.035
22 -0.293 -0.194 -0.098
23 0.111 -0.038 0.152
24 0.228 -0.005 0.234
25 0.130 0.079 0.054 0.060 0.000 0.060
26 0.147 0.051 0.099
27 -0.145 -0.050 -0.096
28 0.257 0.205 0.046
29 0.121 0.127 -0.009
30 -0.022 0.105 -0.128
31 0.098 0.017 0.082 0.080 0.080 -0.010

The first column is the day: 1 is the 1st of the month. For example, 13th is the 13th of january. “Close to close” is the average from the previous days close to todays close. For example, The 1st is the gain from the previous day until the close of the 1st. “Close to open” is the average gain from yesterdays close to todays open. “Open to close” is the average gain intraday from open to the close. The three columns called “average” is simply the average of the 5 days prior to the row.

There are two main patterns that have held up remarkably well over the years: The best days tend to cluster around the end and the beginning of each month.

Of course, this is a very crude analysis, but later I’ll show some potential strategies based on this.