Building a comprehensive set of Technical Indicators in Python for quantitative trading

09 Dec Building a comprehensive set of Technical Indicators in Python for quantitative trading

The model discovers the hidden structure, patterns, or clusters in the data and provides insights or recommendations based on the data. This type of learning can be used for various tasks in technical analysis, such as clustering and dimensionality reduction. Clustering is the task of grouping similar inputs together based on their features, such as price movements, volatility, or trading volume. Dimensionality reduction is the task of reducing the number of features or variables in the data while preserving essential information or variation. Common unsupervised learning algorithms for technical analysis include k-means, hierarchical clustering, principal component analysis, and t-distributed stochastic neighbor embedding.

Whether you’re an experienced trader looking to automate your trading strategies or a beginner interested in learning quantitative trading, this book will be a valuable resource. As defined above, a slow ATR represents 5 days moving average and fast ATR represents 15 days moving average. Using pandas datareader for Yahoo finance database, I extract daily Apple and Netflix stock data from January 1990 to today.

Trend-Following Strategy

The LSTM algorithm has the ability to store historical information and is widely used in stock price prediction (Heaton et al. 2016). The research has successfully developed and tested a stock price prediction model that uses a combination of technical analysis techniques and machine learning methods. The results show that the proposed model is more accurate than traditional methods, especially in the face of high market volatility. The findings emphasize that the integration between technical data and machine learning algorithms not only improves prediction accuracy, but also provides a deeper understanding of stock market dynamics. Although time series and large language models operate under different dynamics (Tan et al. 2024), conventional methods of deep learning often underperform in complicated and noise-filled environments such as the stock market.

Parameters tuning

In addition, data from that year is generally complete and includes various trends and relevant economic events, thus providing a more representative picture of stock market dynamics in regular situations. Rule-based machine learning (RBML) is a branch of machine learning that automatically discovers and learns ‘rules’ from data. It provides interpretable models, making it useful for decision-making in fields like healthcare, fraud detection, and cybersecurity. Key RBML techniques includes learning classifier systems,99 association rule learning,100 artificial immune systems,101 and other similar models. Despite these challenges, our proposed CNN-based model demonstrates better performance than the constant price method.

Integration of Technical Analysis and Machine Learning to Improve Stock Price Prediction Accuracy

At the top, the time series graph of the closing price provides an overview of the trend and pattern of price movements over time. Meanwhile, the ACF and PACF graphs at the bottom show machine learning technical analysis the correlation between the current price and prices at various lags. The slowly decreasing ACF pattern suggests an element of dependence or persistence in the stock price, which is often found in financial data. Then Figure 5 shows the trend and stationarity in stock prices through a scatter plot comparing today’s closing price with the previous day’s closing price.

• Generalisations of Bayesian networks that can represent and solve decision problems under uncertainty are called influence diagrams.
• Our analysis reveals that the most prominent studies regarding LSTMs and DNNs predictors for stock market forecasting create a false positive.
• Besides, most ARIMA models take the previous 10 days or less as input data (Dhyani 2020); this short time would not be enough to capture the more complicated dynamics of the stock market, which can last for a period of at least one fiscal quarter (3 months).
• Another practical implication is the potential use of the model for various investment scenarios, such as day trading, where stock price volatility is higher and faster and more accurate predictions are required.

Approaches

An alternative is to discover such features or representations through examination, without relying on explicit algorithms. The blue line shows the actual closing prices of stock number 2 during a 130 days period. This line is similar to the diagrams presented in (Noel 2023; Pang et al. 2020; Phuoc et al. 2024; Guangyu Ding 2020; Gülmez 2023).

• Tensor Processing Units (TPUs) are specialised hardware accelerators developed by Google specifically for machine learning workloads.
• Due to its generality, the field is studied in many other disciplines, such as game theory, control theory, operations research, information theory, simulation-based optimisation, multi-agent systems, swarm intelligence, statistics and genetic algorithms.
• The trend component reflects long-term price movements, while the seasonal component shows patterns that repeat at certain intervals.
• A recurring question in this domain is whether there exists any reliable technical rule that works consistently over time, or a pattern in the prices that can guarantee long-term profit during predictable windows.
• Further research exploring the model in international stock markets or in various economic conditions will provide additional insights and can improve the reliability and generalization of the prediction model.

Therefore, we recommend using significantly larger datasets (e.g., more than 1000 stock tickers) for each training instance to increase generalizability. The first, fundamental analysis, is concerned with evaluating a company’s financial statements and broader economic indicators to determine the intrinsic value of a security. The result of such an analysis aims to provide the true worth of investment based on factors such as the company’s financial health, the market demands, growth prospects, and prevailing economic conditions. Investors perform fundamental analysis to decide whether to invest in a company based on its current and projected value (Anon. n.d.; Graham 1949). This approach to the market often allows the analysts to look beyond investors’ preferences and the firm’s marketing, foreseeing the company’s potential for long-term success.

When apply LSTM algorithm and technical analysis indicators to forecast price trends on the Vietnamese stock market. In this section, authors will present the results of the data after performing the analysis according to the research process and method, as well as make comments and discuss the research results. Another practical implication is the potential use of the model for various investment scenarios, such as day trading, where stock price volatility is higher and faster and more accurate predictions are required. By identifying the strengths and limitations of this prediction model, this research provides a stronger basis for the development of more accurate and applicable prediction tools in everyday investment practice.

The Long Short Term Memory (LSTM) algorithm introduced by the research of Hochreiter and Schmidhuber (1997) aims to provide better performance by solving the Gradient Vanishing problem that repeated networks will suffer when dealing with long strings of data. In LSTM, each neuron is a “memory cell” that connects previous information to the current task. The LSTM can capture the error, so that it can be moved back through the layers over time. LSTM keeps the error at a certain maximum constant, so the LSTM network can take a long time to train, and opens the door to setting the correction of parameters in the algorithm (Liu et al. 2018).

Basically, technical analysts believe that based on stock prices and the pattern extracted from them, they can access the same information that fundamental analysts derive from news, earnings reports, and annual revenue. However, technical analysts hold the advantage of swift reaction to the market, as their decision-making process can happen on a daily basis. To the authors’ knowledge, no extensive study has been conducted to determine whether technical methods yield better results than fundamental methods. Many of the rules employed by technical analysts come from the advantage of hindsight.

The author uses Microsoft Office Excel software and Python language to calculate technical analysis indicators, process and analyze data. The Long Short Term Memory (LSTM) model is built on the basis of the Sklearn, Keras and Tensorflow support libraries. Similar to Simple Moving Average of price, a simple moving average of volume provides insights into the strength of signal that the stock displays. This chart visualizes how the stock price has moved in relation to its short-term (20-day) and medium-term (50-day) trends. Let us assume that we are currently on 31st December 2018 and have created the model files.

What are the most important machine learning techniques for a Technical Analyst to know?

However, testing all 54 indicators at once would take a long time and a personal computer may not have the power for this request. We recommend using 5 indicators, which results in 31 possible combinations (i.e., each indicator by itself, all 5 indicators together, and all combinations of 2, 3, and 4 indicators). By default, the app scales X features with StandardScaler(), but within the app, the user can test out different data scaling methods. To determine the initial values that will be given to GridSearchCV, upon which it will work to find the best combination, we can use Validation Curves for each of the parameters. Validation curves also look at cross validation and provides a score of prediction for in sample and out of sample. This provides us a good idea of the initial value around which we can provide a range to the GridSearchCV.

In this manner, most time series can be identified if a model large enough is utilized. Besides, some seasonal (long-term) factors can be added to the model to include specific time-windows’ effects (Anon. n.d.). This book will introduce you to the fundamental concepts of quantitative trading and shows how to use Python and popular libraries to build trading models and strategies from scratch. Furthermore, evaluation metrics such as RMSE (Root Mean Square Error) and MAE (Mean Absolute Error) provide quantitative measures of the model’s predictive accuracy in forecasting stock prices. Figure 6 illustrates the autocorrelation and partial autocorrelation (ACF and PACF) for the closing stock price.