•In this research we designed and optimized artificial intelligence (AI) trading systems for intraday trading of five precious metals.•We adopt two technical tools that are known to be useful trading ...stocks, to short term commodities price trends detecting. Relative strength index and keltner channels.•We optimized the setting of our system using particle swam optimization (PSO) which helped us to conduct complex optimization with multiple objectives and under many constraints' variables.•We find that the RSI system outperformed the B&H returns for Gold, Silver, platinum and palladium and was beaten by the B&H returns for copper trades. The system has delivered 106.2%, 63.7%, 22.4% and 326.3% excess returns for Gold, Silver, platinum and palladium.•Both RSI and KC AI systems have been proven to be able to trade profitably precious metals with both long and short positions, in most cases the system performed better in for long trades than for short trades.
In this research we designed and optimized Artificial Intelligence (AI) trading systems for intraday trading of five precious metals. We used data from the beginning of 2020 till the end of September 2021 to design and optimize trading systems using Relative Strength Index (RSI) and Keltner Channels (KC) oscillators. Our prime optimization tool was Particle Swam Optimization (PSO) which helped us to conduct complex optimization with multiple objectives and under many constraints' variables. We find that the RSI system outperformed the B&H returns for Gold, Silver, Platinum and Palladium and was beaten by the B&H returns for Copper trades. The system has delivered 106.2%, 63.7%, 22.4% and 326.3% excess returns for Gold, Silver, Platinum and Palladium. Sixty minutes bars with 1.5 Average True rang Multiplier (MATR) have been found to be a fruitful configuration for the KC system trading Gold, Silver and Palladium providing better trading returns than the B&H strategy, by 64.72%, 58.5% and 310.25%, respectively. Both RSI and KC AI systems have been proven to be able to trade profitably precious metals with both long and short positions, in most cases the system performed better in for long trades than for short trades.
Computational intelligence in finance has been a very popular topic for both academia and financial industry in the last few decades. Numerous studies have been published resulting in various models. ...Meanwhile, within the Machine Learning (ML) field, Deep Learning (DL) started getting a lot of attention recently, mostly due to its outperformance over the classical models. Lots of different implementations of DL exist today, and the broad interest is continuing. Finance is one particular area where DL models started getting traction, however, the playfield is wide open, a lot of research opportunities still exist. In this paper, we tried to provide a state-of-the-art snapshot of the developed DL models for financial applications. We not only categorized the works according to their intended subfield in finance but also analyzed them based on their DL models. In addition, we also aimed at identifying possible future implementations and highlighted the pathway for the ongoing research within the field.
•All searchable articles of deep learning (DL) for financial applications are reviewed.•DL for finance studies based on their application areas were clustered.•DL models according to their performances in different implementation areas were compared.•To best of our knowledge, this is the first comprehensive DL survey for financial applications.•Current status of DL in finance was provided, also the future opportunities were highlighted.
•We converted 1-D financial technical analysis data to 2-D images for classification.•We used 2-D deep convolutional neural network for trend forecasting.•We propose a robust algorithmic trading ...model that works in any market condition.•To best of our knowledge, 2-D CNN with TA has not been used for financial trading before.•Model outperformed Buy & Hold, RSI, MA, LSTM, MLP over long time periods.
Computational intelligence techniques for financial trading systems have always been quite popular. In the last decade, deep learning models start getting more attention, especially within the image processing community. In this study, we propose a novel algorithmic trading model CNN-TA using a 2-D convolutional neural network based on image processing properties. In order to convert financial time series into 2-D images, 15 different technical indicators each with different parameter selections are utilized. Each indicator instance generates data for a 15 day period. As a result, 15 × 15 sized 2-D images are constructed. Each image is then labeled as Buy, Sell or Hold depending on the hills and valleys of the original time series. The results indicate that when compared with the Buy & Hold Strategy and other common trading systems over a long out-of-sample period, the trained model provides better results for stocks and ETFs.
This research paper presents a deep learning-based predictive model for classifying currency trends using technical indicators. The model is trained on a dataset generated from three technical ...indicators: relative strength index (RSI), moving average convergence divergence (MACD), and stochastic. The dataset consists of historical currency data along with the corresponding values of the technical indicators. The deep learning model can accurately classify the trends of a given currency based on the importance of these indicators. The model's performance is evaluated using standard metrics, and the results demonstrate its effectiveness in classifying currency trends. The proposed model provides a valuable tool for traders and investors in the foreign exchange market by helping them make informed decisions about the direction of currency prices.
Research on algorithmic trading using reinforcement learning has become increasingly popular in recent years. Although most of the current reinforcement learning methods are employed to train the ...agent for some kind of modeling or data problem, it is worthwhile to explore in aligning agents with human behavior in applications as crucial as financial trading. Achieving such consistency by incorporating human expert experience into agent behavior is a key for potential improvements in this field. Imitation learning learns directly from examples of humans or other agents performing tasks. However, using imitation learning alone suffers from the problem of transitionally fitting expert example data. By combining the advantages of imitation learning and the Advantage Actor–Critic method, the Human Alignment Advantage Actor–Critic (HA3C) algorithm is proposed, to enhance single-asset trading strategy. First, by adding daily and weekly frequency trading data as input features to TimesNet, which is specifically designed to extract correlated temporal patterns from time-series data, it can capture both short-term and long-term features, thus capturing time-series features more comprehensively. Second, an expert action labeling method is proposed to train a strategy prediction network through supervised learning of behavior imitation. Third, a pre-trained strategy network is transferred to balance the exploration and exploitation of the agent’s behavior. Imitation learning techniques leverage finance-specific knowledge to enhance algorithmic trading consistency. This approach enables algorithms to mimic and adapt human decision-making patterns in finance, ultimately improving overall performance. This paper introduces a novel return-based function that efficiently balances short-term and long-term returns over flexible time horizons. It considers the maximum return from different positions and uses flexible time windows to capture trends while maximizing returns. Finally, evaluation on six commonly used datasets, such as DJI and SP500, demonstrates the advantages of the proposed HA3C algorithm compared with other classical and reinforcement learning-based strategies. Notably, on the HSI dataset, the HA3C strategy significantly outperforms other methods, achieving an impressive cumulative return of 681.55% and a Sharpe ratio of 5.07. These results show the superior performance of the HA3C algorithm in enhancing stock trading strategies and its potential to impact algorithmic trading consistency through aligning agent behavior with human expertise.
•The proposed HA3C method integrates reinforcement and imitation learning.•TimesNet network combines daily and weekly data and excels in expert policy.•A novel reward function maximizes returns with short/long-term balance.•HA3C converges faster and performs better with pre-trained expert policy network.•HA3C significantly outperforms the classical/DRL strategies on six datasets.
Markets are different now, transformed by technology and high frequency trading. In this paper, I investigate the implications of these changes for high frequency market microstructure (HFT). I ...describe the new high frequency world, with a particular focus on how HFT affects the strategies of traders and markets. I discuss some of the gaps that arise when thinking about microstructure research issues in the high frequency world. I suggest that, like everything else in the markets, research must also change to reflect the new realities of the high frequency world. I propose some topics for this new research agenda in high frequency market microstructure.
•With rise in algorithmic trading efficiency the trade size decreases significantly.•Algorithmic trading efficiency leads to decline in trade sizes for small-cap stocks.•Our study uses direct ...identification of Algorithmic Trading instead of proxy.•The trade size decline indicates improved liquidity.•Results encourage the use of Algorithmic Trading in emerging order-driven markets.
Financial markets have come across a phenomenal adoption of advanced and complex technologies in the pursuit of efficient markets. Algorithmic Trading (AT) is one of the prominent moves in this direction and is widely adopted across world markets. The existing literature on AT and its impact on markets is still in the nascent stage primarily due to the inability of most of the markets to directly identify AT. In this study, we directly identify AT and examine its impact on trade sizes which has a key impact on liquidity and price impact of trades. We also use the inverse of Order-to-Trade (1/OTR) ratio as a measure of algorithmic trading efficiency and examine its relationship with size. It is expected that AT has the capability to break large orders into smaller sizes in order to access liquidity and reduce price impact. In this study, we provide empirical evidence for the size effects of AT with direct identification of AT.
•Reinforcement learning (RL) formalization of the algorithmic trading problem.•Novel trading strategy based on deep reinforcement learning (DRL), denominated TDQN.•Rigorous performance assessment ...methodology for algorithmic trading.•TDQN algorithm delivers promising results surpassing benchmark strategies.
This scientific research paper presents an innovative approach based on deep reinforcement learning (DRL) to solve the algorithmic trading problem of determining the optimal trading position at any point in time during a trading activity in the stock market. It proposes a novel DRL trading policy so as to maximise the resulting Sharpe ratio performance indicator on a broad range of stock markets. Denominated the Trading Deep Q-Network algorithm (TDQN), this new DRL approach is inspired from the popular DQN algorithm and significantly adapted to the specific algorithmic trading problem at hand. The training of the resulting reinforcement learning (RL) agent is entirely based on the generation of artificial trajectories from a limited set of stock market historical data. In order to objectively assess the performance of trading strategies, the research paper also proposes a novel, more rigorous performance assessment methodology. Following this new performance assessment approach, promising results are reported for the TDQN algorithm.
•Algorithmic trading using self-attention based recurrent reinforcement learning is developed.•Self-attention layer reallocates temporal weights in the sequence of temporal embedding.•Hybrid loss ...feature is incorporated to have predictive and reconstructive power.
Algorithmic trading based on machine learning has the advantage of using intrinsic features and embedded causality in complex stock price time series. We propose a novel algorithmic trading model based on recurrent reinforcement learning, optimized for making consecutive trading signals. This paper elaborates on how temporal features from complex observation are optimally extracted to maximize the expected rewards of the reinforcement learning model. Our model incorporates the hybrid learning loss to allow sequences of hidden features for reinforcement learning to contain the original state’s characteristics fully. The self-attention mechanism is also introduced to our model for learning the temporal importance of the hidden representation series, which helps the reinforcement learning model to be aware of temporal dependence for its decision-making. In this paper, we verify the effectiveness of proposed model using some major market indices and the representative stocks in each sector of S&P500. The augmented structure that we propose has a significant dominance on trading performance. Our proposed model, self-attention based deep direct recurrent reinforcement learning with hybrid loss (SA-DDR-HL), shows superior performance over well-known baseline benchmark models, including machine learning and time series models.
Fintech, Distristributed Ledgers Tecnology (DLT), blockchain, machine learning, algorithmic trading and High Frequency Trading (HFT), are among the most disruptive digital innovations that are ...transforming the structure of any industrial sector, including the financial industry. Together with the positive spillovers of the introduction of these new technologies (i.e. reducing transaction costs, reducing operating costs, improving speed and security of the transactions, …), we should be aware of the potential new risks that may involve the financial system, whose activity is guaranteed by the trust of the operators. Regulators and supervisors should therefore extend their understanding of the new technologies, both to assess their potential impact on banks' business models and to address risks arising with due caution. Similarly, banks operating within the new technological framework, must rethink their own business models and consider the upcoming challenges, which require specific knowledge and skills.