Web Analytics Made Easy - StatCounter
 
  • Stat Oasis Team

End Of Month Pattern

Updated: Sep 19

Many traders spend a lot of time sifting through blogs, watching videos, reading financial news just trying to make the right decision on when to invest their hard earned money.

Unfortunately most of them either leave a lot of returns on the table, or even worse lose their money trying to time the market or picking the best stocks, sector, or index.


A lot of prominent traders started to push people to index investing to get rid of timing the market and just stay invested in the total market. Buy and hold strategy has been "successful" since 1990 returning on average about 8% per year with many 5-30% drawdowns and couple of +50% shocks.


I think there are many ways to invest better without having to become an Algorithmic Trader.


I have built systems to time the market successfully with RSI2 indicator based on Larry Connors research.



I also built many portfolios that trade once a month with better performance than buy & hold. You can view videos about the subject here:





Today I will show you another very popular timing module which depends on the phenomenon that most stocks, hence their indexes tends to go up at the end of every month. The advantage of trading this pattern, is there are fundamental reasons behind it, and it is very easy to implement with much lower volatility than buy & hold.

it has been working for the past 125 years and still do till today because of the size behind it, and it doesn't look its going away anytime soon.


Lakonishok and Smidt (1988) were the first to have reported a turn-of-the-month seasonality in stocks returns in their white paper “Are seasonal anomalies real? A ninety-year perspective”. The researchers have found that, on average, the four days at the turn-of-the-month account for all of the positive returns to the DJIA over the period of 1897-1986.


Another white paper “Calendar Anomalies in Stock Index Futures“ written by Carcano and Tornero, support this phenomenon “Our analysis reveals that the turn-of-the-month effect in S&P 500 futures contracts is the only calendar effect that is statistically and economically significant and persistent over time".


Since then a lot of white papers have been written on the subject with many (hedge funds) using the study as the basis for their system design with billions of dollars invested in the phenomenon directly or through derivatives.


Fundamentals behind the pattern

  • Most researchers attribute this effect to the timing of monthly cash flows received by pension funds, which are reinvested in the stock market. The end of the month is also a natural point for portfolio/trading models rebalancing both for retail and professional investors.

  • A systematic monthly shift in interest rates does not appear to explain the turn-of-the-month pattern in equity returns.

  • Caution is needed if one implements this strategy as calendar effects tend to vanish or rotate to different days in a month.


Takeaways from the papers:

  • Virtually all of the excess market return is accrued during the turn of the month (4 days) period, and investors received a small or no reward for bearing the market risk over the other 16 trading days of the month. Remember there are 20 trading days in a month on average.

  • Despite the fact that the effect is more pronounced among small-cap and low-price stocks, it also exists for large-cap and high-price stocks.

  • The turn-of-the month effect is not concentrated at calendar-year quarter-ends.

  • The reason for functionality also is not a risk-based; Using the standard deviation of return as a measure of risk, it was found that risk is not higher during the four turn-of-the-month days than over the other 16 trading days of the month.

  • Interestingly, the turn-of-the-month effect occurs in 30 different markets, so we can conclude that the effect is not due to a factor unique to the U.S. market structure.

Conclusions:

In spite of the simplicity the turn of the month is a well-known phenomenon on stock indexes and other markets, with many ways to implement the idea into a system.

for example, buy SPY ETF two trading days before the end of the month and to sell it on the second trading day of the upcoming month, it is both profitable and statistically significant.


Trading day of the month is not the same as calendar day of the month. While months have 30 calendar days on average, they have about 20 trading days on average. That has to be taken care of when programming the system.


Implementing the End of Month Strategy:

Here are the strategy rules:

Instruments (Futures): SP500, Midcap 400, Dow Jones, Nasdaq, Russell 2000

Data: Daily timeframe starting January 1997 till August 2022

Size: Fixed one contract per trade per instrument

Commission & Slippage: $0

Long Entry Signal: Buy (x) trading days before end of month

Long Exit Signal: Sell (y) trading days after start of month

Filters: None


I combined all above instruments in a portfolio and tested 100 combinations starting at trading day 14 to 23 (step 1) and exiting on next month trading day 1 to 10 (step 1).

Each run has +1200 trades.



Out of the 100 optimizations, 80% of combinations were profitable as a portfolio since 1997, the chunk of data includes many market regimes, including the dot com bubble with +50% drawdown in SP500 and +85% drawdown in Nasdaq, the subprime bubble again with +50% drawdown and many 5-30% drawdowns in between.


Below you can see the top 25% of combination had an average trade of +$500 and average net profit of +645,000 which is amazing.




The best run made $837,537 with an average trade of $648


Interestingly, September & June show losses in most combination, so it would make sense to filter them out when you are building your system




Further Research:

  • You can easily add a filter to enhance multiple metrics in this portfolio

  • You can focus on subset of Russell index to push returns higher

  • You can filter out September & June to enhance portfolio returns

  • If you trade stocks or ETFs, you can build a portfolio of portfolios by trading multiple systems on same instruments:

  • System 1: Buy 3 trading days before end of month, sell 2nd trading day of month

  • System 2: Buy 5 trading days before end of month, sell 5th trading day of month

  • System 3: Buy 8 trading days before end of month, sell 7th trading day of month

  • The advantage of combining the three systems above into a portfolio is to smooth out the equity curve.



You can download the strategy code from the discord server:

https://Go.StatOasis.com/Discord


I recently posted a video with a strategy based on the same pattern





248 views0 comments

Recent Posts

See All