Technical analysis strategy optimization using a machine learning approach in stock market indices

A comparative machine learning technical analysis of state-of-the-art of deep learning model for stock market prediction. Between 2010 and 2017, they used 424 S&P 500 Index Component Stocks and 185 CSI 300 Index Component Stocks , comparing traditional machine learning algorithms with advanced DNN models. Traditional ML algorithms involve Logistic Regression, Random Forest, Classification and Regression Tree , SVM, and eXtreme Gradient Boosting, whereas DNN architectures contain MLP, Deep Belief Network , GRU, RNN and LSTM. Their results indicate that in many of the directional assessment indicators conventional ML algorithms have higher performance without knowing transaction costs, but DNN models show better results given transaction costs. It has been widely studied and explored for the prediction of stock price direction. Machine learning tasks are usually classified as supervised and unsupervised learning.

Predicting direction of stock price movement is notably important to provide a better guidance to assist market participants in making their investment decisions. This study presents a hybrid method combining machine learning methods with persistent homology to improve the prediction performance. Three stock prices namely Kuala Lumpur Composite Index, Kuala Lumpur Stock Exchange Industrial and Kuala Lumpur Stock Exchange Technology sampled from Kuala Lumpur Stock Exchange are selected for experimental evaluation. For comparative analysis, we compare the proposed method with others, where the machine learning methods are applied independently on stock returns and also on technical indicators respectively.

J. Econ. Finance Adm. Sci.

This progression of computations through the network is called forward propagation. The input and output layers of a deep neural network are calledvisiblelayers. The input layer is where the deep learning model ingests the data for processing, and the output layer is where the final prediction or classification is made. As shown in Table 4, the best metric to evaluate LSTM testing set is RMSE with 0.002 of error percentage. In addition, the best result of error percentage for R2 is 0.537 with the features set of (High/Low/Open/Close) and 250 epochs.

Bollinger band is essentially an average price of a security and its 95% confidence interval which means 95% of the times the security price remains inside this band. Suppose we have the following plot demonstrating the price of a stock over time. Coupled with supporting momentum, a crossover might constitute a strong buy or sell signal.

Rate of change is a momentum indicator that explains a price momentum relative to a price fixed period before. RSI is one of the most common momentum indicator aimed at quantifies price changes and the speed of such change. A major point discussed was their realisation of the importance of time dependency in graphical representation as to whether or not time-dependent signals can be detected in static images. Images were labelled according to trading opportunities, and trading volume information was incorporated into the chart’s candlesticks — reminiscent of TrendSpider’s Raindrops. Lastly, we will look at a research paper conducted by Naftali Cohen, Tucker Balch and Manuela Veloso of J.P. This paper covered computer vision for stock chart analysis on the S&P 500.

The experimental results are presented in Section 5, and finally, we draw the main conclusions and discuss possible future developments in Section 6. In a more recent work, the adaptive markets hypothesis has been proposed to overcome the behavioral critics made to EMH arguing that markets are not rational, but are rather driven by fear and greed. AMH tackles the stock market from a biological perspective within an evolutionary framework in which prices evolve according to competition, adaptation, and natural selection to financial interactions. According to AMH, predictable patterns may appear over time for short periods. Such difficulties have led to the efficient-market hypothesis , which states that asset prices already take into account the information based both on past and future events.

Applied Artificial Intelligence

Gaussian Mixture is an uses a probabilistic method of determining the appropriate cluster for a series of observation, assuming the universe is formed out of different Gaussian distributions. The idea is to use returns of these companies and create an elbow curve to determine the number of clusters that would create a balance between low sum of squares within a cluster vs the total number of clusters. For 100 companies, if we create 100 clusters, we would get 0 sum of squares within the cluster however this clustering would not make sense, i.e. would not be parsimonious. On the other hand, having 1 cluster would be parsimonious however will lead to very high sum of squares within cluster.

• I find that the new index exhibits statistically and economically significant in-sample and out-of-sample predictive power and outperforms the well-known technical indicators and macroeconomic variables.
• As well as, we used the same features set and the number of epochs to calculate the model error.
• For a trading horizon of days, technical factors may have some value; in general, the longer the trading horizon, the less valuable technical factors are.
• One way to solve this problem is to create different ML model for each cluster of companies that are expected to behave similarly perhaps belonging to the same industry, where the “behavior” is captured in their returns.
• Correspondingly, we also value decision complexity less at a trading horizon of this size.

Among the multitude of methods used to predict this movement, technical indicators have been around for quite some time as one of the methods used in forming an opinion of a potential move. Support vector machines have been proven to be useful for regression analysis and forecasting. When stochastic uncertainty is involved in the datasets, robust support vector regression models are useful.

A hybrid model integrating deep learning with investor sentiment analysis for stock price prediction

It’s also best to avoid looking at machine learning as a solution in search of a problem, Shulman said. Some companies might end up trying to backport machine learning into a business use. Instead of starting with a focus on technology, businesses should start with a focus on a business problem or customer need that could be met with machine learning. From manufacturing to retail and banking to bakeries, even legacy companies are using machine learning to unlock new value or boost efficiency. For the first event, we see the price crossing from the outside to the inside of the upper Bollinger Band.

This is done so that https://trading-market.org/ly guessing a class label will result in a 33% success rate. Last 1000 periods of data was used because they represent the current market climate better than older data. Twint was used to simplify the tweet webscraping process instead of Twitter’s API due to the API’s limited nature. Time series data was downloaded as daily data then resampled into weekly and monthly intervals. Deep learning requires a great deal of computing power, which raises concerns about its economic and environmental sustainability. “The more layers you have, the more potential you have for doing complex things well,” Malone said.

However, neural networks is actually a sub-field of machine learning, and deep learning is a sub-field of neural networks. SMA calculates the average of prices over a given interval of time and is used to determine the trend of the stock. To provide Machine Learning algorithms with already engineered factors, one can also use (SMA_15/SMA_5) or (SMA_15 – SMA_5) as a factor to capture the relationship between these two moving averages. Regarding the choice of machine learning algorithms, computer vision stands out as an immediately intuitive option. Traditionally, technical analysis is a visual endeavour in which a human trader manually identifies and classifies patterns on a visual chart. Thus, it makes sense that object detection and classification models would be well suited to doing the same, likely with a higher accuracy than a human user.

A variety of layers, epochs, and nodes were experimented with to find the optimum amount. •Develop a workflow to calculate the performance of the proposed strategy and compare it with technical analysis-based strategies. Much of the technology behind self-driving cars is based on machine learning, deep learning in particular. This pervasive and powerful form of artificial intelligence is changing every industry. Here’s what you need to know about the potential and limitations of machine learning and how it’s being used.

• Similar to Simple Moving Average of price, a simple moving average of volume provides insights into the strength of signal that the stock displays.
• Most of the indicators are created using Close rather than Adjusted Close in this article.
• The energy industry isn’t going away, but the source of energy is shifting from a fuel economy to an electric one.
• NLTK’s VADER then analyzes the sentiment which is later visualized with the donut chart and histogram chart.

The black line represents the actual stock prices, while the red line represents the predicted stock prices. According to the chart above, we can notice that there are high volatility periods, even though we have low volatility periods. Thus, every period of high volatility started with a decline in returns. Similarly for the RNN model, we compare the results we obtained from both training and testing sets with the same types of evaluation metrics. As well as, we used the same features set and the number of epochs to calculate the model error. Machine learning and deep learning models are capable of different types of learning as well, which are usually categorized as supervised learning, unsupervised learning, and reinforcement learning.

A sequence of successful outcomes will be reinforced to develop the best recommendation or policy for a given problem. Semi-supervised learning offers a happy medium between supervised and unsupervised learning. During training, it uses a smaller labeled data set to guide classification and feature extraction from a larger, unlabeled data set. Semi-supervised learning can solve the problem of not having enough labeled data for a supervised learning algorithm. Since technical indicators work best in short term, I will use 5 days and 15 days as my fast and slow signal respectively. The following indicators are customizable to any duration with a single parameter change.

•Study the return of the technical analysis-based strategies TEMA and MACD. Madry pointed out another example in which a machine learning algorithm examining X-rays seemed to outperform physicians. But it turned out the algorithm was correlating results with the machines that took the image, not necessarily the image itself. Tuberculosis is more common in developing countries, which tend to have older machines. The machine learning program learned that if the X-ray was taken on an older machine, the patient was more likely to have tuberculosis.

According to EMH, it is not possible to predict future prices based on historical data since for such purpose it is necessary to possess privileged information. Some critics to EMH point to the psychological biases that investors exhibit under uncertainty, leading to irrational and unpredictable behaviors . Nowadays there is no consensus about EMH and the debate is still ongoing.

Many companies are deploying online chatbots, in which customers or clients don’t speak to humans, but instead interact with a machine. These algorithms use machine learning and natural language processing, with the bots learning from records of past conversations to come up with appropriate responses. The moving average of price and the percent change in volume consider only price and volume, respectively, so they are both technical indicators. Understand the structure and techniques used in machine learning, deep learning, and reinforcement learning strategies. Reinforcement machine learning is a machine learning model that is similar to supervised learning, but the algorithm isn’t trained using sample data.