Sell in May and Go Away?

You may or may not have heard the expression “Sell in May and go away, come back again after Labor day” in regards to the stock market.  There has long been the belief that the positive returns in the stock market display seasonal tendencies which occur between September through May with the summer months offering lack luster returns.

The theory is that if you sell your stocks in May and move to all cash for the summer that on average you will outperform the overall market for the year.

I have taken historical data on the Dow Jones industrial index dating back to 1915 to see how this strategy would have performed over the long run vs. just buy and hold.   I have assumed that you would earn 4% interest on any cash balance in the account, which in today’s day and age of ultra low interest rates, doesn’t hold true, but over the course of the entire test period, it should be sufficient.

The results for the baseline buy and hold strategy appear below:

Starting Equity: $10,000
Ending Equity: $2,044,264
Total Return: +20,342.64%
Avg. Annual Return: +5.76%
Maximum Drawdown: -89.48%

The rules for the initial sell in May system are pretty straightforward.  Buy the Dow Jones index the first trading day in September and sell the index the first trading day in May, moving to all cash.

The results are as follows:

Starting Equity: $10,000
Ending Equity: $733,346
Total Return: +7,236.46%
Avg. Annual Return: +4.56%
Maximum Drawdown: -87.85%

Based on these results you would have done MUCH worse selling in May and buying back in September than if you just held through the summer.  The buy and hold strategy returns +1.20% more per year, which seems like a small amount, but due to the time value of money over the extremely long test period you would have missed out on more than $1.2 million in profits.   Not only this, but one would hope the volatility of the account may be reduced in order to compensate for the reduced profitability, but this is not the case when looking at the maximum drawdown of the 2 strategies,  -89.48% for buy and hold vs. -87.85% for sell in May.

Clearly based on these results selling in May is not a good strategy to implement, but it may be possible to make this strategy better by adding an additional timing filter.   The timing filter we will look is to only buy the index after the first trading day in September AND only after the index has closed above its X day moving average.  If the index still isn’t above its X day moving average by the end of the year, then buy the first trading day in January.   Conversely, we only sell after the first trading day in May after the index has crossed below its Y day moving average.  If the index is still above its Y day moving average by the end  of July, then sell the index the first trading day in August.  The X and Y values will be optimized over the test period.

The graph of the optimization of the X and Y values appears below.

Optimization Results

As can be seen, there are a large range of values for which x and y will work for this strategy.   I selected an X value of 130 and a Y value of 120 with the results listed below:

Starting Equity: $10,000
Ending Equity: $8,751,504
Total Return: +87,415.05%
Avg. Annual Return: +7.29%
Maximum Drawdown: -68.31%

Adding an additional timing filter to the sell in May strategy not only improves the performance above the baseline sell in May strategy, but it also provides an improvement over the buy and hold strategy by a significant margin, while at the same time helping to reduce the account volatility by bringing the maximum drawdown of the account down to -68.31%.

The Sell in May and Go Away adage may have a nice ring to it, but over the long term you would be better off  just buying and holding.  By applying additional filters however, this could be a feasible strategy to trade in a long term investment account.

web: http://LimitUpFutures.collective2.com
twitter: @LimitUpTrading

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s