TRADERS´ English I November 2013 - page 22

INSIGHTS
22
11.2013
The
Halloween
Effect
The Halloween effect states that stocks should be purchased at the end of October and sold at the beginning of May. For all
those who have their doubts about this market anomaly, Prof Jacobsen of Massey University has published a study proving
beyond any doubt that such an anomaly does exist. This article gives a short overview of the results as well as of further
improving approaches.
Good Times ahead? – New Evidence
Thomas Hupp
Thomas Hupp, along with Michael Uhl, is the
initiator of the MSF 4D asset oscillator fund of
funds and an adjunct professor of Asset Man-
agement at Loerrach University of Cooperative
Education in Germany. Since 1996, he has been
studying investment strategies, focusing –
since 2008 – on seasonal effects of inefficient
stock markets.
in 35 of these countries the anomaly is statistically
significant.
The scientific rigour applied to achieve such results is
outstanding and beyond any criticism. Even when the time
factor is taken into account, the Halloween effect stands
up. To prove this, Jacobsen examined both 300-and 60-
year periods as well as rolling 2.5 and 10-year periods
to find out that in 82 per cent of all cases, the Halloween
effect generated significantly higher returns than a Buy-
and-Hold strategy. These findings, in particular the insight
into which stock markets are suitable for a Halloween
strategy, are helpful for investors (see Figure 1).
Selecting markets may be simple: Just take ten to 16
significant countries from the Jacobsen study, find the
appropriate ETFs and buy these at the time of entry. The
timing may be simple, too: Buy the ETFs or certificates on
31st October and sell them on 1st May.
Regarding timing we conducted a long term study
covering a period of 40 years that suggested that the
month of July was a better time to exit (see TRADERS´
07/2013). Then, we carried out quite a few studies using
indicators to find out the best times for entry and exit.
Here the results generated by technical analysis were the
more successful ones.
To determine how fund selection and technical
analysis operate together, we have combined both results
in a simulation.
Bouman, S, & Jacobsen, B ( 2002):
The Halloween Indicator, Sell in May and go Away: Another
Puzzle
Jacobsen, B, & Zhang, C (2012 ):
The Halloween Indicator: Everywhere and All the Time
Sources
»
Using all available share prices data, he analysed 108
stock markets in different places and over different
time periods. The upshot was that there is such a thing
as the Halloween effect in nearly every stock market
in the world. In 81 countries returns are higher from
November to April than during the rest of the year, and
1...,12,13,14,15,16,17,18,19,20,21 23,24,25,26,27,28,29,30,31,32,...80
Powered by FlippingBook