Limits...
Integrated model of multiple kernel learning and differential evolution for EUR/USD trading.

Deng S, Sakurai A - ScientificWorldJournal (2014)

Bottom Line: Initially, the prediction model optimized by MKL predicts the returns based on a technical indicator called the moving average convergence and divergence.Next, a combined trading signal is optimized by DE using the inputs from the prediction model and technical indicator RSI obtained from multiple timeframes.The experimental results showed that trading using the prediction learned by MKL yielded consistent profits.

View Article: PubMed Central - PubMed

Affiliation: Graduate School of Science and Technology, Keio University, 3-14-1 Hiyoshi, Kohoku-ku, Yokohama 223-8522, Japan.

ABSTRACT
Currency trading is an important area for individual investors, government policy decisions, and organization investments. In this study, we propose a hybrid approach referred to as MKL-DE, which combines multiple kernel learning (MKL) with differential evolution (DE) for trading a currency pair. MKL is used to learn a model that predicts changes in the target currency pair, whereas DE is used to generate the buy and sell signals for the target currency pair based on the relative strength index (RSI), while it is also combined with MKL as a trading signal. The new hybrid implementation is applied to EUR/USD trading, which is the most traded foreign exchange (FX) currency pair. MKL is essential for utilizing information from multiple information sources and DE is essential for formulating a trading rule based on a mixture of discrete structures and continuous parameters. Initially, the prediction model optimized by MKL predicts the returns based on a technical indicator called the moving average convergence and divergence. Next, a combined trading signal is optimized by DE using the inputs from the prediction model and technical indicator RSI obtained from multiple timeframes. The experimental results showed that trading using the prediction learned by MKL yielded consistent profits.

Show MeSH
Example showing the relative strength index values from multiple timeframes.
© Copyright Policy - open-access
Related In: Results  -  Collection


getmorefigures.php?uid=PMC4109363&req=5

fig1: Example showing the relative strength index values from multiple timeframes.

Mentions: Our trading time horizon is 1 hour, which means that we assess overbought or oversold signals based only on 1-hour time frame data. Clearly, it is possible that the judgment would be different if we made assessments using a longer or shorter timeframe. For example, Figure 1 shows the EUR/USD rate and its RSI values for 1-hour and 2-hour timeframes (i.e., 1-hour RSI and 2-hour RSI values). Note that at the eighth point (10:00:00, May 5, 2011) in Figure 1, the 1-hour RSI value is approximately 73.90, which provides us with a sell signal because the currency pair is overbought, whereas the 2-hour RSI value is approximately 43.98, which tells us that the currency is not overbought. The rate increased further from the eighth to the ninth point (11:00:00, May 5, 2011). In addition, the 1-hour RSI value is approximately 78.32 at the ninth point and the 2-hour RSI value is approximately 71.71, which suggests that both values provide overbought signals so it is highly probable that the rate will decrease from the ninth point onwards. This example shows that if we use the RSI to generate trading rules, we must assess the overbought or oversold conditions not only for the target timeframe, but also for relatively longer and shorter timeframes. For example, the features of the RSI from a relatively shorter timeframe (i.e., 30 minutes in this study) and a relatively longer timeframe (i.e., 2 hours) were used in this study as suitable signals for trading a target currency pair.


Integrated model of multiple kernel learning and differential evolution for EUR/USD trading.

Deng S, Sakurai A - ScientificWorldJournal (2014)

Example showing the relative strength index values from multiple timeframes.
© Copyright Policy - open-access
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC4109363&req=5

fig1: Example showing the relative strength index values from multiple timeframes.
Mentions: Our trading time horizon is 1 hour, which means that we assess overbought or oversold signals based only on 1-hour time frame data. Clearly, it is possible that the judgment would be different if we made assessments using a longer or shorter timeframe. For example, Figure 1 shows the EUR/USD rate and its RSI values for 1-hour and 2-hour timeframes (i.e., 1-hour RSI and 2-hour RSI values). Note that at the eighth point (10:00:00, May 5, 2011) in Figure 1, the 1-hour RSI value is approximately 73.90, which provides us with a sell signal because the currency pair is overbought, whereas the 2-hour RSI value is approximately 43.98, which tells us that the currency is not overbought. The rate increased further from the eighth to the ninth point (11:00:00, May 5, 2011). In addition, the 1-hour RSI value is approximately 78.32 at the ninth point and the 2-hour RSI value is approximately 71.71, which suggests that both values provide overbought signals so it is highly probable that the rate will decrease from the ninth point onwards. This example shows that if we use the RSI to generate trading rules, we must assess the overbought or oversold conditions not only for the target timeframe, but also for relatively longer and shorter timeframes. For example, the features of the RSI from a relatively shorter timeframe (i.e., 30 minutes in this study) and a relatively longer timeframe (i.e., 2 hours) were used in this study as suitable signals for trading a target currency pair.

Bottom Line: Initially, the prediction model optimized by MKL predicts the returns based on a technical indicator called the moving average convergence and divergence.Next, a combined trading signal is optimized by DE using the inputs from the prediction model and technical indicator RSI obtained from multiple timeframes.The experimental results showed that trading using the prediction learned by MKL yielded consistent profits.

View Article: PubMed Central - PubMed

Affiliation: Graduate School of Science and Technology, Keio University, 3-14-1 Hiyoshi, Kohoku-ku, Yokohama 223-8522, Japan.

ABSTRACT
Currency trading is an important area for individual investors, government policy decisions, and organization investments. In this study, we propose a hybrid approach referred to as MKL-DE, which combines multiple kernel learning (MKL) with differential evolution (DE) for trading a currency pair. MKL is used to learn a model that predicts changes in the target currency pair, whereas DE is used to generate the buy and sell signals for the target currency pair based on the relative strength index (RSI), while it is also combined with MKL as a trading signal. The new hybrid implementation is applied to EUR/USD trading, which is the most traded foreign exchange (FX) currency pair. MKL is essential for utilizing information from multiple information sources and DE is essential for formulating a trading rule based on a mixture of discrete structures and continuous parameters. Initially, the prediction model optimized by MKL predicts the returns based on a technical indicator called the moving average convergence and divergence. Next, a combined trading signal is optimized by DE using the inputs from the prediction model and technical indicator RSI obtained from multiple timeframes. The experimental results showed that trading using the prediction learned by MKL yielded consistent profits.

Show MeSH