Efficient Gap Forex Trading Strategy

I came across a scientific investigation.

It has been intensively engaged in the analysis of gaps. Different markets were tested:

  • commodities (oil and gold)
  • US stock market (Dow Jones Index and the IBM share)
  • Russian stock market (MICEX and the share Sberbank)
  • FOREX (EUR/USD, GBP/USD and USD/RUB)

By applying various statistical tests, it has been shown that in a large number of cases the observed price behaviour is not inconsistent with market efficiency.

The market efficiency hypothesis (Fama, 1970) states that prices already take full account of existing information and are therefore random.

Therefore, it should not be possible to make systematic gains from past behaviour. However, there have already been some studies that could refute this.

This study also reveals another anomaly that refutes the hypothesis of market efficiency.

In this case, the trading strategy generated high profits by exploiting this anomaly.

Causes of Gaps

The six most common explanations for gaps are as follows:

  • Unexpected events, such as earnings or other important news items
  • Drastic changes in market conditions
  • Post-trade developments
  • Considerable time delays between the previous closing price and the current opening price (caused by weekends and holidays)
  • Technical reasons, such as a significant expansion of the bid-ask spread
  • Other reasons

Hypotheses tested

The period tested was from 2000-2015 and thus 16 years. The following hypotheses (H) were tested:

  • H1: Prices tend to rise after upward gaps;
  • H2: Prices tend to fall after downward gaps;
  • H3: Prices tend to rise before upward gaps;
  • H4: Prices tend to fall before downward gaps;
  • H5: Gaps are of short duration
  • H6: Profits resulting from gaps are different from normal profits

H1 and H2 deal with the price behaviour after the occurrence of gaps.

H3 and H4 deal with whether the occurrence of gaps is predictable or not.

H5 checks the statement that all gaps will be closed.

H6 examines whether gaps are an anomaly that is in conflict with market efficiency or not.

Empirical evaluations

Since the focus of this article is on FX trading, I will first only go into more detail about the empirical evaluations regarding the tested FX pairs.

Furthermore, it has also been shown that there is insufficient evidence in the other markets for the existence of an anomaly of gaps that contradict market efficiency.

However, in the case of the Dow Jones Index, in summary, the probability of a further price increase in the event of an upward gap was 80%. From this, one could derive a strategy that opens a long position after an upward gap in the Dow Jones and closes it again at the end of the day.

In the case of the IBM share, the best strategy would have been to shorten after an upward or downward gap and close the position again at the end of the day.

In the case of the Sberbank share, the price dynamics only differed from the normal price dynamics in the event of downward gaps. A profitable trading strategy for the tested period would be to open a short position after a downward gap and close the position at the end of the day.

In FX, it was easy to see that the price dynamics in gaps were clearly different from the normal price dynamics. Especially after upward gaps.

This behavior was best seen in the EUR/USD and GBP/USD currency pairs.

Due to the negative regression after a gap opening in these forex pairs, the following strategy was tested more closely with regard to its profitability:

Sell EUR/USD and GBP/USD after a gap opening and close the position at the end of the day.

An example of an upside gap and shortening the gap at the opening. At the end of the day, the trade is closed again.
Thus, a trading robot was developed that traded different gap sizes.

The parameters were 0.05% – 1% and the steps were 0.05%.

It turned out that although the total profit was slightly higher for the EUR/USD at a gap size of 0.05%, the drawdown was almost twice as high.

Therefore, a gap size of 0.10% was chosen for the EUR/USD. For the GBP/USD, a gap size of 0.05% provided the best overall result.

The results of the backtesting of the trading strategy in EURUSD and GBPUSD

As can be seen from the table, profits can be considered stable.

The average hit rate is 60% and higher. In EUR/USD the hit rate was even 100% in 2006 and 2014.

Losses occurred in only 3 out of 16 years in EUR/USD, and in only 2 out of 16 years in GBP/USD.

The z-tests also clearly show that the results obtained are statistically different from the random ones.

The hypothesis of market efficiency was thus refuted in both cases, revealing a simple but nevertheless profitable anomaly.

Conclusion

The investigation shows that trading does not always have to be complicated. You just have to be able to identify anomalies that work and generate performance despite their “simplicity”.

Moreover, such approaches are also very easy to program yourself and are especially useful for automated trading.

It is also a good example of the fact that not many trades per year are necessary to achieve a certain performance.

Nevertheless, I would not adopt this approach for my own trading. But you can use this anomaly as a basis to create an interesting system.

Quants and system traders do nothing else in principle. They look for systems that have a statistical advantage in the long run and let them run fully automated. These are much more complex, but the procedure is the same.


Read my other articles about Forex Trading:

Leave a Reply

Your email address will not be published. Required fields are marked *