Radiqant



Split Time Series in Two Semesters

In this analysis we will consider the majors FX futures: AUD, CAD, CHF, EUR, GBP, JPY, NZD. The goal is to identify correlations that support Contrarian Investing, an investment strategy characterized by buying and selling against the prevailing sentiment.

More specifically, in this study we are going to identify FX futures that had a negative performance in the first half of the year, and then see if their value appreciated in the second half of the year. 

Table 1 summarizes through the Win Ratio, Avarage Tick Movements, and the number of times negative performance occurred the first half of the year, what happened in the second half of the year.

Fx future

Win Ratio

Average tick movements

Total trade

AUD

0,40

0,0409

5

CAD

0,50

0,0059

6

CHF

0,80

0,0471

5

EUR

0,57

0,0340

7

GBP

0,67

0,0344

6

JPY

0,40

-0,0002

5

NZD

1,00

0,0491

5

Table 1 –The table summarizes through the Win Ratio, the average return and the number of times a negative performance occurred in the first semester, what happened in the second semester.

As can be seen from the table, 5 out of 7 futures have a Win Ratio of 50% or more. 

In total, out of 39 observations, 24 semesters close positive after performing neagtively the first semester, with a Win Ratio of 62%.

Only 1 out of 7 futures (JPY) has a negative Average Yield.

In total, the 39 semesters observed produce avarage tick movements of 0.0302

This investment strategy gains strength and relevance from the good quantitative data that result and support it.

Note that this analysis based on contrarian investing is set up in its simplest form: it can be improved by making links to further investigations.

Giuseppe Ferrulli

CEO, Radiqant

Share on

Other articles

TREND FOLLOWINGIn this analysis we will consider the AUD futures. Dividing the time series starting in 2003, we will look for an entry that breaks a monthly high-open candle formed the same month as the year before.
Mean Reversion In this analysis, we will consider the CAD futures. Dividing the time series starting in 2002, we will look for the existence of a correlation between the various months.