I like to take a quantitative approach to the markets and I find seasonal factors very interesting. In this article I investigate whether trading on different days of the month can deliver better returns in the stock market.
I will firstly look at the returns on each trading day of each month. Then I will turn this into a trading strategy that exploits this information and compares it to a buy and hold strategy.
I was listening to an excellent Podcast Interview with Jay Kaeppel. In this interview Jay talks a lot about seasonal factors and I wanted to have a look and see whether I could replicate his results.
Market and Data
I am carrying out this test on the S&P 500 index. The S&P 500 is weighted by capitalization and represents the share price of the largest US listed companies. The data is from Yahoo Finance and I have used a 21 year period from the start of 1995 to the end of 2015.
My analysis has been carried out using a Tradinformed Excel Backtest Model. These are an affordable and flexible way for anyone to test their own trading strategies on any market and timeframe.
Trading Day Results
The results showed that there is a significant variation between the returns on each trading day. The 1st trading day of each month was the most profitable day to be in the market. This is closely followed by the returns on the 11th trading day.
Trading Day Trading Strategy
Based on the returns by trading day I have analyzed a trading strategy that is in the market on:
- Days 1-3
- Days 9-12
- Days 19 – end of the month
Comparing Trading Day Strategy with Buy and Hold
I tested the Buy and Hold and Trading Day Strategy and I got the following results.
|Buy and Hold||Day of the Month Strategy|
|Gross Winning Trades||$250,558||$1,752,846|
|Gross Losing Trades||0||$-1,157,348|
|Largest Winning Trade||$250,558||$40,751|
|Largest Losing Trade||0||$-32,836|
|Average Profit Per Trade||$250,558||$1,233|
|Number of Trading Years||21||21|
|Annual Compounded Return||6.20%||9.70%|
The results showed that the Trading Days Strategy was more than twice as profitable over this time period. In addition the maximum drawdown is less than half of the Buy and Hold Strategy. Looking at the equity curves below we can see that the Buy and Hold experienced deep declines in 2000-2002 in the Dot-com Bubble burst and 2008-2009 Financial Crisis.
The results above show that the Trading Day Strategy significantly outperforms Buy and Hold based on historic data. However, the above test does not take account of trading costs and these would reduce the profitability of the strategy. On balance I think these results could form a useful trading strategy, particularly if combined with other seasonal factors or technical analysis.
If you are interested in trading strategies there are lots more on the Trading Strategies page.