Seasonalities: The Near-Term Future for the Market

Sell in May and go away. Is there any truth to this? Did some work on seasonalities recently and applied it to the stock market to quantify the truthfulness of this statement.

The video recording is available from my YouTube channel.

First the methodology. To discover seasonalities, I decided to split each year in equal (or approximately equal) periods. The most natural period is a single month, and that’s what I started with. Later, I went a bit further by dividing each month into two. Thus, 24 periods per year, of approximately the same length.

Then, I run a python script to pull out some statistics for the various single periods (Dow Jones Industrial Average, 30 years of history):

Notice the highlighted unusual weakness during June and August. The statistics over 15 years of history are similar:

Thus, the summer months are often weak. Indeed, my script failed to find any multi-period sequences which has been strong historically including these months (June, July or August). The conclusion: the summer is indeed weak.

Interestingly enough, based on similar analysis we are currently in a strong period for the stock market. Here is an example:

Mid-March until end of April has been strong historically. Apparently what has been developing this year is a deviation, unless the market rallies into the end of the month.

All in all, the summer weakness is there, and if history is to repeat – there might be further troubles for the stock market. Combine that with the recent Dow Theory sell signal, and the situation starts to look gloomy. At least for the near term future.