Extracting or predicting future stock price trends from the current and historical stock trading activity is an open research problem. Convolution Neural Network (CNN) recently shown excellent performance to extract high-level features from raw data in many domains. In this research work, we trained CNN to extract future stock price trends. Trend Following is a trading strategy that does not require prior knowledge of the future stock price trend. In the Trend Following trading strategy, traders buy stock in an uptrend and sell the stock in a downtrend. Trend Following trading strategy is based on the current price trend and the assumption that the current price trend will continue further for some time, but if the current price trend does not proceed then this trading strategy fails to provide the profit. Our objective is to improve the Trend Following trading strategy by combining the advantages of both the CNN-based prediction and the Trend Following trading strategy. In this research work, we proposed a modified CNN-based Trend Following trading strategy named Deep Trend Following (DTF) in which trading decisions are based on the current and predicted future price trend of the stock. Our experimental results are twofold. Firstly our trained CNN-based classifier outperformed the baseline methods. Secondly, the DTF trading strategy outperformed three trading strategies viz the CNN-based, Buy-and-Hold, and the simple Trend Following trading strategies on the American and the Indian index stocks.
Stock market trading is a fascinating domain in search of lucrative profit. The movement of the stock price is non-linear, non-stationary, and noisy as the stock price is based on various factors, and most of these factors are uncertain. Many factors can move the stock price up or down including demand-and-supply of stock, company’s management, competition from competitors, government policy, the policy of the central bank of the country, any news related to the company can influence the stock price.
market is eficient, and it reflects the efect of information on the stock price.
Traders and investors buy, sell, or hold (do nothing) stocks for profit, but their objectives are diferent. The perspective of the investor is long term based on the long term trend and the prospect of the company, having a stockholding period ranging from a year to a few years.
In contrast, traders buy (sell) and sell (buy) the stock
based on the short term trend, having a stockholding
period of a few days, hours, minutes, or seconds. The
investors predict the long term trend of stock based on
MUFin21 (International Workshop on Modelling Uncertainty in the
nEvelop-O
downtrend([
Extracting or predicting future stock price trends from gorithm. In ([
The prediction or classification of stock price or trend performance evaluation of the proposed trading system. using machine learning methods is looking promising Concluding remarks with suggestions for future research but is not always reliable. So, taking a trading decision are given in Section 5. based only on the prediction of future price trends is not a good strategy. The TF strategy based solely on the current stock price trend also does not look promising. 2. Methodology The combination of these two strategies will reduce the disadvantages of both approaches. Our objective is to This section provides a concise presentation to Convoluimprove the TF trading strategy by combining the advan- tional neural networks. tages of both, i.e. CNN-based future stock price trends prediction and TF trading strategy. In this research work, 2.1. Convolutional Neural Network we proposed a modified CNN-based TF trading strategy, The typical architecture of CNN consists of a series of and we name it as Deep Trend Following (DTF) in which layers, and these layers can classify as the Input layer, trading decisions will base on the current price trend and Convolutional layer, Pooling layer, Fully connected layer, CNN-based predicted future price trend of the stock. and Output layer as shown in Figure 1 ([12, 13]) ([14]).
We experimented with the proposed DTF strategy on the American and the Indian index stocks. As we created the DTF strategy using a CNN-based binary classifier to predict the next session, + 1 close price trend and simple Trend Following trading strategy. Our experimental results are twofold, firstly we evaluated the performance of our CNN-based classifier with the baseline method on the same American index stocks dataset used in the baseline paper ([4]) and the results reported in the Table 2 and Table 3. Secondly, we evaluated the performance of the proposed DTF strategy with the CNN-based, Buyand-Hold, and simple Trend Following trading strategies using performance evaluation metrics mentioned in Section 4.4 on the American and the Indian index stocks and the results reported in Table 4.
In ([4]), authors use CNN to predict the next-day stock
price trend using features from diverse sources. They Figure 1: Convolutional Neural Network
represented the input data in 2D and 3D forms. In ([
Trend Following is a stock trading strategy that takes 2.1.1. Convolutional layer trading decisions based on the current stock price trend. The first layer in a CNN is a convolutional layer, as shown It takes a trading decision whenever current stock price in Figure 1, this layer extracts features called a feature trend changes, i.e. from an uptrend to downtrend or map from the input data by performing the convolu- downtrend to uptrend for that it continuously observes tion operation using a filter. In this work, input data the current trend. In stock market trading buy first and represented as a two-dimensional array (, ) having sell later is called a Long position. Sell first and buy then number of rows and number of columns. A two- is called a Short position. Traders take a long position dimensional filter (, ) having number of rows and whenever they think the stock price will increase and number of columns used. The convolution operation take a short position whenever they feel the stock price performed on two functions so the input and filter will decrease. can be thought as functions, the convolution operation on these two functions performed as per the Equation 1 3. Proposed Trading System ([15]) to produce another function which is a 2-D feature map having − +1 rows and − +1 columns. The values in the filter learned while training the network.
Predicting the future stock price trend and optimal utilization of the prediction along with implicit provision /2 /2 for risk management is the fundamental idea of this work. (, ) = (, )⊗ (, ) = ∑ ∑ (+, +)∗ (, ) In this section, we propose the complete end-to-end Al=−/2 =−/2 gorithmic trading strategy using the Trend Following
(1) strategy and CNN. Our proposed trading system will
The activation function is applied to the output of work on single security (or stock) for simplicity and will the convolution operation as shown in Equation 2. The take only one trading decision, i.e. buy, sell, or hold (do nonlinear activation function commonly used is Relu nothing) a previous trading position on every trading () = max(0, ) . session. We have included the transaction cost of 0.10% of the trading amount for buy or sell in our study. We /2 /2 have performed experimentation on the daily stock data (, ) = ( ∑ ∑ ( + , + ) ∗ (, ) ) (2) having open, high, low, and close price information along =−/2 =−/2 with the volume of the trading.
The proposed trading system is shown in Figure 2, 2.1.2. Pooling layer in which the CNN module is responsible for predicting The pooling layer reduces the dimensionality of the fea- future stock price trends. It has trained from historical ture map by subsampling of the data while retaining stock market data. It will predict the next trading session the useful features. It reduces the computational cost ( + 1) close price trend from live and historical stock and also it is a measure to avoid overfitting of the CNN market data. The technical analysis indicators module model. In pooling operation, pooling window used, the tells the current () stock price trend at the closing time of values under the window reduced to a single value, which the current session from live and historical stock market reduces the size of the input to the next layer. The com- data. monly used pooling operation is the max pooling, in The Trend Following trading agent will take a tradwhich among the values, the maximum value selected. ing decision at the open time of the next trading session + 1 . It is based on the prediction of CNN about the 2.1.3. Fully Connected Layer stock close price trend of next trading session ( + 1) and current () stock price trend provided by technical The convolution operation, activation function, and pool- analysis indicators as shown in the Algorithm 1. In the ing operation are responsible for extracting features from Algorithm 1, CR_trend denotes the current () stock price the input. To map these extracted features in the pre- trend provided by the technical analysis indicators, and vious layers into the final output, at the end of CNN, NX_trend indicates the prediction of CNN about the next a fully connected network, i.e. multilayer perceptron trading session ( + 1) stock close price trend. We are tak(MLP) is appended. It converts the last 2-D feature map ing a trading decision at the opening of the next trading to the 1-D feature vector. In a classification problem, a session + 1 instead of the close of the current trading single output selected from the feature vector based on session because non-trading period information also the probabilities of all the outputs using a probabilistic plays a significant role to decide the stock price trend. function. Softmax is the commonly used probabilistic function for this task.
This section describes the data representation to train the CNN model. The detail of the baseline algorithm to compare the performance of the proposed classifier. The experimental setup of the proposed DTF trading strategy, 4.2. Baseline algorithm also about performance evaluation metrics used. Detail of the CNN-based, Buy-and-Hold, and simple Trend Following trading strategies used to compare the performance of the proposed DTF strategy and also discusses the experimental results. Our proposed Deep Trend Following trading strategy uses the stock price trend predicted by the two-dimensional CNN, to make a trading decision. CNN predicts the next session stock price trend based on the current and historical stock trading data, so input data to CNN is represented in the two-dimensional matrix form, as shown in Figure 3. We have derived features from the stock trading data, and these total features form the columns of the matrix. The entire rows of the matrix are the current and historical days, i.e. trading sessions. The optimal value of is ten, i.e. ten recent days’ data to predict the next trading session close price selected by experimentation on diferent values of . The features used and the calculations are described in Table 5.
Total features are thirty, which can group as technical indicators, simple moving averages, exponential moving averages, day of the week. Day of the week is a very significant feature because the global stock markets correlated up to a certain extend. The non-trading hours period is also a significant factor in the stock price trend.
The experimental data is divided into training and testing, as shown in Table 1.
The training data is very crucial for the performance of the machine learning model. We identified the outlier in our dataset based on the observation of the daily % changes in the close price, as shown in Figure 4. We removed the outlier using Z-statistics.
Our baseline algorithm is the one reported in ”Ehsan Hoseinzade and Saman Haratizadeh (2019)”. We compare the performance of the proposed method with the baseline algorithm ([4]). We have experimented with the proposed method with the same dataset mentioned in the baseline algorithm. Compared the results of the proposed method with the results reported in the baseline algorithm. Experimental Data of the American index stocks DJIA, NASDAQ, S&P 500, RUSSEL and Indian index stock NIFTY Index Stock
We have used % Accumulated Return, % Maximum
Drawdown, % Average Daily Return, % Average Annual Return,
Sharpe Ratio, Standard Deviation, Skewness, and
Kurtosis[
We compare the performance of our proposed DTF trading strategy with the Simple Trend Following, Buy-andHold, and CNN-based trading strategy in terms of % accumulated return. In the CNN-based trading strategy buy action is taken when CNN predicts, the price will go up at the close of the next trading session + 1 , and it will take sell action when CNN predicts the price will go down. Buy action suggestion ignored if an already long position is open similarly sell action ignored if an already short position is open.
Experimentation in this work is grouped into two tasks.
In the first task, we proposed CNN based binary classifier and compared the results of the classification with the classifiers in the baseline paper. In the second task, we introduced the DTF trading strategy using CNN based binary classifier and simple Trend Following trading strategy. Also, compared the result of the proposed DTF Derived 82 features from data to train the CNN model. 4.5. Trading Strategies
Our experimentation grouped into two tasks. First, we predicted the next day’s stock price trend at the close price of the next day trading session, this is a classification problem, as shown in the equation below.
= { 1, 0, if else _ +1 > _
We compare the classification results with the classification result of the baseline method and the other methods reported in the baseline paper, as shown in Table 2 and Table 3 ([4]).
Second, as per our proposed method, we created the DTF trading strategy using classification result of the proposed classifier and Trend Following trading strategy.
We compared the result of the DTF trading strategy with trading strategy with the buy-and-hold, CNN-based, and 5.4. This comparison is shown in Table 4. The % Accusimple Trend Following trading strategy. mulated return of the proposed DTF trading strategy
In the first task, as mentioned in Section 5.3, we com- is considerably better than the other trading strategies
pare the classification result of the proposed CNN-based mentioned in Table 4. The comparison of all four trading
binary classifier with the baseline classifiers in terms of strategies including proposed DTF trading strategy in
the macro-average and macro-best F-measure scores. We terms of % Accumulated return for index stocks DJIA,
used the macro-average-F-measure comparison metric NASDAQ, S&P 500 and NIFTY is shown in Figure 5,
Figinstead of accuracy because the financial data has an ure 6, Figure 7 and Figure 8 respectively. Figure 9,
summaimbalanced class distribution [
The proposed classifier outperformed the baseline clas- datasets. It is better than simple Trend Following trading sifiers and other classifiers reported in the baseline paper strategy on two datasets and approximately equal for the in terms of the average and best F-measure score. Our other two datasets. The proposed DTF method’s % Daily classifier outperformed the baseline classifier because a and % Annual Returns is better than all diferent trading deep learning method required extensive training data, strategies, as shown in Table 4. Similarly, the Sharpe and we used large training data compared to the baseline ratio of the trading strategy should be maximum, and method. Also, Their input matrix has 60 rows, i.e. 60 the Sharpe ratio of the proposed DTF method is better previous days, and we have used ten past days. Only than all other trading strategies, as shown in Table 4. recently trading days influence the stock price trend, and The skewness and kurtosis of the proposed DTF method they used too many last days’ data in the input matrix. on all datasets are reasonably acceptable. The volatility, We also used the feature to capture non-trading time i.e. standard deviation of the all four trading strategy, is information. approximately the same.
In the second task, the performance of the proposed DTF trading strategy compared with the performance of the CNN-based, simple Trend Following, and buy-andhold trading strategies on the American index stock DJIA, NASDAQ and S&P 500; the Indian index stock NIFTY using performance evaluation metrics mentioned in Section 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30.
Close price Open price Week day RSI R CCI RSI trend R trend CCI MOM1 MOM2 MOM3 VC1 VC2 VC3 OHC OLC OCC LHC SMA5 SMA10 SMA15 SMA20 SMA25 EMA5 EMA10 EMA15 EMA20 EMA25 NT
Description Daily close price of the stock Daily open price of the stock Day of the week Relative Strength Index Williams Percent Range (%R) Commodity Channel Index if RSI>70 then 2, if RSI<30 then 1 else 0, if R>-100 then 2, if R>100 then 1 else 0, if CCI<70 then 2, if CCI<30 then 1 else 0, One day % return Two days % return Three days % return % change in volume compare to volume one day before % change in volume compare to volume two days before % change in volume compare to volume three days before % change between open and high price % change between open and low price % change between open and close price % change between low and high price Simple moving average n=5 days close price Simple moving average n=10 days close price Simple moving average n=15 days close price Simple moving average n=20 days close price Simple moving average n=25 days close price Exponential Moving Average n=5 days close price Exponential Moving Average n=10 days close price Exponential Moving Average n=15 days close price Exponential Moving Average n=20 days close price Exponential Moving Average n=25 days close price Non trading period information Source/Formula Yahoo Finance Yahoo Finance Datetime Library TA Library TALibrary TA Library [( [( ℎ ℎ [( [( [( [( [ EMA EMA EMA EMA EMA [(
−1 − − − −
− − prediction is risky. The performance of the Trend Follow- compared the results with the CNN-Based, simple Trend can not predict the future stock price trend. In this paper, strategies in terms of percentage Accumulated Return,
Owing to the noisy and non-stationary behavior of the stock price trend, the trading strategy based only on the ing trading strategy depends on the stock price trend after trading action is taken but the Trend Following strategy we proposed a modified Trend Following trading stratwhich is a combination of CNN-based prediction with egy named Deep Trend Following (DTF) trading strategy, ratio.
The main contribution of this research work is that it combines the CNN-based prediction with the Trend Following strategy in which the Trend Following strategy will take a trading decision based on the current trend and future trend predicted by the CNN-based classifier. This combination reduces the risk based on the only prediction. We experimented with the proposed DTF trading strategy on the American and the Indian index stocks and Following, and Buy-and-Hold trading strategy. The proposed DTF model outperformed the other three trading Maximum Drawdown, Average return, and the Sharpe
We also compared our CNN-based binary classifier with the baseline CNN-based binary classifier, and it outperformed the baseline classifier in terms of macroaverage and macro-best F-measure scores. It performed better because we have considered the non-trading period information, and we have taken more training data [15] G. S. Tandel, M. Biswas, O. G. Kakde, A. Tiwari, H. S.
Suri, M. Turk, J. R. Laird, C. K. Asare, A. A. Ankrah, N. Khanna, et al., A review on a deep learning perspective in brain cancer classification, Cancers 11 (2019) 111.