Revisiting the Halloween Effect

Revisiting the Halloween Effect Author  – Acting-Man

From Crash Danger to End-of-the-Year Ramp

 

[Ed note by PT: we are unfortunately a week late in posting this issue of SI, which didn’t reach us in time due to a technical problem. We decided to post it belatedly anyway: for one thing, the effect under discussion is normally in effect until the end of the year; for another, the statistical validity of this information goes beyond the current year, as it is a recurring phenomenon. Lastly we would note that we have a strong suspicion that the effect may actually be cut short this year – details to be discussed in a follow-up post]

 

Knock, knock… it’s Jack, from Wall Street

 

After the last day of October, the month that has investors in fear of the next big stock market crash follows Halloween  – and while this is a spooky day, it actually marks the start of a period that typically tends to yield promising returns for investors.

What is commonly referred to as Halloween Effect or Halloween Strategy is the fact that stock market returns on average turn noticeably positive from late October onward.

This phenomenon has been widely discussed well beyond the confines of experts on seasonality like ourselves and was inter alia subject of several academic studies, with a number of well-known scholars providing valuable contributions (Jacobsen & Visaltanachoti; Haggard & Witte; Maberly & Pierce and many others)*.

We are revisiting the topic in this edition of Seasonal Insights.

 

The Halloween Effect is a Global Phenomenon

We begin by looking at the first part of the Halloween Effect with the help of the  Seasonax Web App (note: details on the web app can be found here; readers of Acting Man qualify for a special discount)

We have analyzed three of the most important stock indexes from around the world to find out whether they exhibit a common pattern. In all three cases the time period from October 31st  to January 3rd was reviewed.

Please note that these indexes have been examined with the aim of showing a general trend. One can filter out individual stocks that exhibit the same seasonal patterns, but generate much higher returns.

The first index is the DAX, a composite of the 30 largest German companies. It shows a distinct seasonal pattern in the respective time-period with 77.78% winning trades of and an average annualized return of 34.05%.

 

The seasonal pattern of the DAX is in line with the theory of the Halloween Effect

 

The second index we looked at is the Russel 2000, the small-cap stock market index of the bottom 2,000 stocks in the Russell 3000 Index. It shows 81.25% winning trades with an average annualized return of 24.83%.

 

In the Russel 2000 the Halloween pattern generated 13 winners and just 3 losing years over the past 16 years.

 

The last index we examined is the Japanese Nikkei 225 – a very important index in a global comparison, as it tends to exhibit deviations from general global trends. However, its seasonal trend is also in line with the global Halloween Effect pattern, with 78.57% winning trades and an annualized return of 44.18%.

 

The Nikkei exhibits roughly the same pattern as the other two indexes.

 

The Halloween Effect is a Reliable Seasonal Pattern

This seasonal analysis of three international indexes indicates that the statistical probability that stock prices will rally from October 31st onward is very high (obviously there are occasional exceptions). Depending on investment style, sector and individual stock analysis can be used as an additional filter to this analysis of the general trend.

Continue Reading / Acting-Man>>>

Sharing is caring!