Complex 5-component oscillator as a tool of technical analysis in algorithmic trading

Размещено на http://www.allbest.ru/

COMPLEX 5-COMPONENT OSCILLATOR AS A TOOL OF TECHNICAL ANALYSIS IN ALGORITHMIC TRADING

B.A. Zhalezka

Belorus National Technical University, Minsk, BELARUS

A.O. Stadnik

Godel Technologies Europe Ltd., Minsk, BELARUS

V.A. Siniavskaya

Belorussian State Economic University, Minsk, BELARUS

Abstract

New complex oscillator is suggested for effective algorithmic trading on the stock market. The oscillator is based on the following technical indicators: moving average convergence divergence, parabolic stop and reverse, average directional moving index rating, momentum volume weighted average price. Rule base for the oscillator construction is created. The algorithm of the oscillator caluculation is realised via QuantOffice software. Results of the oscillator backtesting on the historical data are presented.

Introduction

Technical analysis is one of the most popular methods for justifying of the stock exchange decisions in conditions when the price of a security is determined only by demand and supply. For such an analysis, the input is only a time series of the price of the security. Based on it, a price change is predicted in the future and various indicators are calculated that form signals for the trader about the most favorable points for the sale or purchase of securities.

There is a wide variety of technical indicators, described in detail, for example, in books [1 - 3]. When forming a trading strategy, not one, but several indicators are usually used. Their complex is called an oscillator. Taking into account the relationships between them allows trader to make more profitable order on selling or buying the securities than using only one indicator.

The purpose of this article is to form a trading strategy based on a five-component oscillator and verify its efficiency. It is the development of the authors's investigation of computer data analysis and modelling in the sphere of the stock exchange decision making [4].

1. Base technical indicators for the oscillator construction

For construction of the oscillator 5 following indicators were selected [5].

1. Moving Average Convergence Divergence (MACD), which is the difference between exponential moving averages of the short and long trade periods. MACD signals a trend change and indicates the start of a new trend. Price divergence points to the end of the current trend, especially if the MACD is at extreme high or low levels.

2. Volume Weighted Average Price (VWAP), which is a ratio of the volume sold to the total volume sold over a certain period (formula (1)).

, (1)

where Volume_i - quantity of the securities selled in i-th time moment; Price_i - security's price in i-th time moment.

VWAP is often used by investors who seek to be as passive as possible in their activity.

3. Parabolic Stop and Reverse (PSAR), which moves at a higher acceleration than a simple moving average, can change its position from the point of view of the price, is designed to determine trend reversal points and exit points. It is calculated according to the following rules (formula (2)).

, (2)

where xp - maximim price value from the moment of the long position opening; af - acceleration factor, its value increments by step value each time the maximum value of xp is exceeded; High - maximum price in i-th time period; Low - minimum price in i-th time period.

4. Average Directional Moving Index Rating (ADXR), which helps to determine an existance of a price trend. It consists of five lines: ADX, ADXR, AD, + DI, -DI [5].

5. Momentum, which measures price acceleration and deceleration (formula (3)). When the market peaks, the Momentum indicator will rise sharply and then fall, deviating from the ongoing price move.

Similarly, at a local low, Momentum prices will plummet and then start to rise significantly ahead of prices. Both of these situations lead to a discrepancy between the indicator and prices.

(3)

3. The algorithm of the oscillator calculation in Quant Office

Complex oscillator, constructed on the base of 5 abovementioned indicators (MACD, VWAP, PSAR, ADXR, Momentum) was authomatized by means of QuantOffice software. A ready-made trading strategy template called EMA Crossover was used. This template is available in the QuantOffice environment as a test variant of the trading strategy, in which orders are executed immediately after they are placed.

The strategy is a .dll file written in C# that is used by QuantOffice. In order to use these strategies as a kind of shell necessary for the operation of the new oscillator, the standard indicators were replaced with the new oscillator.

A database of rules was developed to generate signals for purchase and sale, as well as to exclude false signals [5], for the implementation of which the following program code was written.

if (Macd.Stable && Vwap.Stable && Sar.Stable && Adxr.Stable && Momentum.Stable)

{

if (Macd.Previous.MACD <= MacdDownThreshold && Macd.MACD > MacdDownThreshold)

{

flacMacdDown = true;

flacMacdUp = false;

}

// buy rules: current price is above the VWAP and ADX is above ADXThreshold and the Momentum is above MomentumUpThreshold and the MACD cross up the MacdUpThreshold the price hits the parabolic to the upside

if (close > Vwap.VWAP && Adxr.ADX > AdxThreshold && Momentum.MomentumValue > MomentumUpThreshold

&& Macd.Previous.MACD <= MacdUpThreshold && Macd.MACD > MacdUpThreshold &&

close < Sar.Previous && close >= Sar.SAR && flacMacdDown)

{

flacMacdDown = false;

result = 1D;

return true;

}

if (Macd.Previous.MACD <= MacdUpThreshold && Macd.MACD > MacdUpThreshold)

{

flacMacdUp = true;

flacMacdDown = false;

}

//sell rules: the price hits the parabolic to the downside and the MACD cross down the MacdDownThreshold and ADX is above AdxThreshold and the Momentum is below MomentumDownThreshold and current price is below the VWAP

if (close > Sar.Previous && close <= Sar.SAR && Macd.Previous.MACD >= MacdDownThreshold

&& Macd.MACD < MacdDownThreshold && Adxr.ADX > AdxThreshold

&& Momentum.MomentumValue < MomentumDownThreshold && close < Vwap.VWAP && flacMacdUp)

{

flacMacdUp = false;

result = -1D;

return true;

}

result = 0D;

return true;

}

4. Oscillator backtesting on the historical data

In the table 1 the results of the developed strategy testing on the historical data of Apple inc. securities (AAPL) are represented.

Table 1. Results of the trade testing on the historical data of AAPL (Apple inc. shares)

By means of new 5-component oscillator 45 transactions were generated, of which 27 (or 60%) were profitable and yielded a net profit of 1,7239.22 cash units. The obtained values of the Sharpe ratio (it is equal to 1.52, the value ?1 is considered as positive) and the Sortino ratio (it is equal to 2.93, the value greater than 2 is considered as effective) indicate that this strategy is effective.

Graphical interpretation of the 5-component oscillator work is represented on the figure 1.

Размещено на http://www.allbest.ru/

Fig. 1 - Graph of the 5-component oscillator signals combined with the price, SAR and VWAP

Conclusion

Automation of the technical analysis allows minimizing the risk of order delaying due to subjective factor of trader's doubting in its efficiency. Algorithmic trading on the base of oscillators is a perspective direction in the sphere of the stock exchange decision making under conditions when only supply and demand can influence on the price of the security. oscillator software algorithmic trading

For the construction of the oscillator 5 technical indicators were used: moving average convergence divergence, parabolic stop and reverse, average directional moving index rating, momentum volume weighted average price. The backtesting of the complex 5-component oscillator allowed concluding that the trading algorithm is effective and gives a profit for its user.

The developed oscillator and trade strategy were approved for use in CJSC Orientsoft - Programs and Solutions (Belarusian division of Deltix inc.).

References

1. Elder, A. (2014) Trading for a Living. Psychology. Trading Tactics. Money Management. John Wiley & Sons, New York. - 303 p.

2. Williams, B., Gregory-Williams J. (2008) Trading Chaos. Maximize Profits with Proven Technical Techniques. John Wiley & Sons, New York. - 258 p.

3. Kaufman, P. (2019) Trading Systems and Methods. John Wiley & Sons, New York. - 1168 с.

4. Siniavskaya, O.A., Zhelezko, B.A. (2007) Fuzzy Evaluation of the Risk of Investment in Securities in the Portfolio Optimization Problem. In: Computer Data Analysis and Modeling: Complex Stochastic Data and Systems: Proc. of the Eighth Intern. Conf., Minsk, Sept. 11-15, 2007. In 2 vol. Vol. 2. Minsk, Publ. center BSU, pp. 91 - 94.

5. Zhalezka, B.A., Stadnik A.O., and Siniauskaya V.A. (2022) Technical analysis and indicators application in algorithmic marketing. Economic Science Today. Vol. 15, pp. 119 - 130. (in Russian).

Размещено на Allbest.ru