Financial Time Series Prediction Research Articles

Multi level perspectives in stock price forecasting: ICE2DE-MDL.

This study proposes a novel hybrid model, called ICE2DE-MDL, integrating secondary decomposition, entropy, machine and deep learning methods to predict a stock closing price. In this context, first of all, the noise contained in the financial time series was eliminated. A denoising method, which utilizes entropy and the two-level ICEEMDAN methodology, is suggested to achieve this. Subsequently, we applied many deep learning and machine learning methods, including long-short term memory (LSTM), LSTM-BN, gated recurrent unit (GRU), and SVR, to the IMFs obtained from the decomposition, classifying them as noiseless. Afterward, the best training method was determined for each IMF. Finally, the proposed model's forecast was obtained by hierarchically combining the prediction results of each IMF. The ICE2DE-MDL model was applied to eight stock market indices and three stock data sets, and the next day's closing price of these stock items was predicted. The results indicate that RMSE values ranged from 0.031 to 0.244, MAE values ranged from 0.026 to 0.144, MAPE values ranged from 0.128 to 0.594, and R-squared values ranged from 0.905 to 0.998 for stock indices and stock forecasts. Furthermore, comparisons were made with various hybrid models proposed within the scope of stock forecasting to evaluate the performance of the ICE2DE-MDL model. Upon comparison, The ICE2DE-MDL model demonstrated superior performance relative to existing models in the literature for both forecasting stock market indices and individual stocks. Additionally, to our knowledge, this study is the first to effectively eliminate noise in stock item data using the concepts of entropy and ICEEMDAN. It is also the second study to apply ICEEMDAN to a financial time series prediction problem.

Enhancing Financial Time Series Prediction with Quantum-Enhanced Synthetic Data Generation: A Case Study on the S&P 500 Using a Quantum Wasserstein Generative Adversarial Network Approach with a Gradient Penalty

This study introduces a novel Quantum Wasserstein Generative Adversarial Network approach with a Gradient Penalty (QWGAN-GP) model that leverages a quantum generator alongside a classical discriminator to synthetically generate time series data. This approach aims to accurately replicate the statistical properties of the S&P 500 index. The synthetic data generated by this model were compared to the original series using various metrics, including Wasserstein distance, Dynamic Time Warping (DTW) distance, and entropy measures, among others. The outcomes demonstrate the model’s robustness, with the generated data exhibiting a high degree of fidelity to the statistical characteristics of the original data. Additionally, this study explores the applicability of the synthetic time series in enhancing prediction models. An LSTM (Long-Short Term Memory)-based model was developed to evaluate the impact of incorporating synthetic data on forecasting accuracy, particularly focusing on general trends and extreme market events. The findings reveal that models trained on a mix of synthetic and real data significantly outperform those trained solely on historical data, improving predictive performance.

Directional Prediction of Financial Time Series Using SVM and Wilson Loop Perceptron

Directional Prediction of Financial Time Series Using SVM and Wilson Loop Perceptron

Exploring Long-Memory Process in the Prediction of Interval-Valued Financial Time Series and Its Application

Exploring Long-Memory Process in the Prediction of Interval-Valued Financial Time Series and Its Application

Analysis of the Relationship Between Cross Capital Flows and Stock Exchange Index with Machine Learning

This paper investigates forecasting the BIST100 stock index using cross-capital flow analysis. It employs feature engineering and the Orthogonal Matching Pursuit (OMP) model to navigate the intricacies of financial time series prediction. The study meticulously selects features such as lagged values, moving averages, and volatility metrics, normalized to ensure unbiased model impact. The OMP model is carefully optimized to handle the dimensionality of financial data, avoiding overfitting through a sparsity constraint. This approach yields an R-squared score of 0.88, indicating a solid capability to capture index variance. Visual comparisons between actual and predicted values further validate the model's accuracy. The paper highlights the importance of methodological precision in developing models capable of discerning complex patterns, offering valuable insights for investment strategies. Implications of the study show that cross-capital movements and macroeconomic variables are a good fit with ML to predict the Stock Market despite the complexity of financial markets.

Application and challenges of informer model in financial time series prediction: A review

Time series prediction has shown excellent performance in various fields in recent years, such as stock prices, weather changes, traffic flow, and other fields. Its application development is becoming increasingly mature, and long series time prediction area of research has gained significant prominence. The excellent performance of deep learning in many models has unleashed the potential and possibility of time series prediction to a certain extent. Based on the above reasons, applying deep learning to the field of time series prediction has become a meaningful research. Therefore, the purpose of this article is to analyze the Informer algorithm model in the area of financial time series prediction and provide a comprehensive literature review on the implementation of Informer models in financial time series. Attempting to investigate and analyze the problems and challenges that Informer models may encounter in the area of financial time series prediction, with the hope of providing innovative inspiration and motivating new forms of knowledge for future workers.

A survey on uncertainty quantification in deep learning for financial time series prediction

Investors make decisions about buying and selling a financial asset based on available information. The traditional approach in Deep Learning when trying to predict the behavior of an asset is to take a price history, train a model, and forecast one single price in the near future. This is called the frequentist perspective. Uncertainty Quantification is an alternative in which models manage a probability distribution for prediction. It provides investors with more information than the traditional frequentist way, so they can consider the risk of making or not making a certain decision. We systematically reviewed the existing literature on Uncertainty Quantification methods in Deep Learning to predict the behavior of financial assets, such as foreign exchange, stock market, cryptocurrencies and others. The article discusses types of model, categories of financial assets, prediction characteristics and types of uncertainty. We found that, in general terms, references focus on price accuracy as a metric, although other metrics, such as trend accuracy, might be more appropriate. Very few authors analyze both epistemic and aleatoric uncertainty, and none analyze in depth how to decouple them. The time period analyzed includes the years 2001 to 2022.

Deep Learning for Financial Time Series Prediction: A State-of-the-Art Review of Standalone and Hybrid Models

Deep Learning for Financial Time Series Prediction: A State-of-the-Art Review of Standalone and Hybrid Models

Time Series Prediction Based on LSTM Neural Network for Gold Prices

Metals like gold have been traded around the world for ages. Gold, the yellow shiny metal, can maintain the stability of the market, disperse the risk of transfer price, and stabilize the national economy. Therefore, the research and prediction of gold prices is not only of great significance to the countries but also of great significance to individuals. However, compared with other financial time series, research on gold prices is still relatively few and lacks models integrated with deep learning. The purpose of this paper is to establish a new model for the time series prediction of gold prices and at the same time to provide new ideas for future financial time series prediction problems. In this paper, a prediction model based on LSTM (long-term memory Neural Network) is proposed by using the time series of large amounts of financial data as experimental data for prediction. The RMSE (root mean square error), which acts as a loss function, is selected to evaluate the accuracy of our prediction model. The model proposed has a good prediction effect on financial time series.

Prediction of US 30‐years‐treasury‐bonds movement and trading entry point using the robust 1DCNN‐BiLSTM‐XGBoost algorithm

AbstractThis article presents a novel algorithm that accurately predicts market trends and identifies trading entry points for US 30‐year Treasury bonds. The proposed method employs a hybrid approach, integrating a 1‐dimensional convolutional neural network (1DCNN), long‐short term memory (LSTM), and XGBoost algorithms. The 1DCNN is used to learn local and short‐term patterns, while LSTM is employed to capture both short and long‐term dependencies. Furthermore, we have implemented an algorithm that utilizes hull moving average (HMA) and simple moving average (SMA) crossover data to detect trading entry points and major trends in the market. The combination of the SMA–HMA crossover algorithm and predictions provided by the 1DCNN‐BiLSTM‐XGBoost algorithm yields exceptional results in terms of prediction accuracy and profitability. Additionally, these integrated techniques effectively filter out noise and mitigate false breakouts, which are often observed with US 30‐year Treasury bonds. In the field of financial time series prediction, the effectiveness of 1DCNN and LSTM in identifying trading entry points and market perturbations has not been comprehensively studied. Therefore, our work fills this gap by demonstrating through experiments that the proposed 1DCNN‐BiLSTM‐XGBoost algorithm, in combination with moving average crossovers, effectively reduces noise and market perturbations. This leads to the precise identification of trading entry points and accurate recognition of trend signals for US 30‐year Treasury bonds. We demonstrate through experiments that our proposed approach achieves an average root mean squared error of 0.0001 and an R‐square value of 0.9999, highlighting its promise as a method for predicting market trends and trading entry points for US 30‐year Treasury bonds.

Multi-step-ahead stock price prediction using recurrent fuzzy neural network and variational mode decomposition

Financial time series prediction has attracted considerable interest from scholars, and several approaches have been developed. Among them, decomposition-based methods have achieved promising results. Most decomposition-based methods approximate a single function, which is insufficient for obtaining accurate results. Moreover, most existing research has concentrated on one-step-ahead forecasting that prevents market investors from making the best decisions for the future. This study proposes two novel multi-step-ahead stock price prediction methods based on different decomposition techniques, including discrete cosine transform (DCT), i.e., a linear transform, and variational mode decomposition (VMD), i.e., a non-linear transform. DCT-MFRFNN, a method based on DCT and multi-functional recurrent fuzzy neural network (MFRFNN), uses DCT to reduce fluctuations in the time series and simplify its structure and MFRFNN to predict the stock price. VMD-MFRFNN, an approach based on VMD and MFRFNN, brings together their advantages. VMD-MFRFNN consists of two phases. The input signal is decomposed to several intrinsic mode functions (IMFs) using VMD in the decomposition phase. In the prediction phase, each IMF is given to a separate MFRFNN for prediction, and predicted signals are summed to reconstruct the output. DCT-MFRFNN and VMD-MFRFNN use the particle swarm optimization (PSO) algorithm to train MFRFNN. In this research, for the first time, the gradient descent method is used to train MFRFNN. Three financial time series are used to evaluate the proposed methods. Experimental results indicate that VMD-MFRFNN surpasses other state-of-the-art methods. VMD-MFRFNN, on average, shows a decrease of 31.8% in RMSE compared to the MEMD-LSTM method. Also, DCT-MFRFNN outperforms MFRFNN and DCT-LSTM in all experiments, which reveals the favorable effect of DCT on MFRFNN’s performance. To assess the effectiveness of PSO in training VMD-MFRFNN, we compared its performance with twelve different metaheuristic approaches. PSO, on average, shows a decrease of 9.4% in MAPE compared to other metaheuristic methods.

Centralized decomposition approach in LSTM for Bitcoin price prediction

It has been reported that integrating time-series decomposition methods and neural network models improves financial time-series prediction performance. Despite its practical importance, the prediction performance of cryptocurrency prices, including Bitcoin, at the tail domain of the label distribution is generally less successful than the mean performance across the entire domain of the label distribution. In order to enhance the overall predictive performance of the Bitcoin price, we propose the Centralized Clusters Distribution (CCD) as a novel input data filtering mechanism that significantly improves both the tail performance and the overall performance by mitigating the extreme bimodality inherent in Bitcoin price. The combination of CCD and the Weighted Empirical Stretching (WES) loss function, which imposes different penalties depending on the label distribution, outcomes in an additional performance gain. In the Long-Short Term Memory (LSTM) and the Singular Spectrum Analysis (SSA) decomposition method, the CCD-WES strategy outperforms the native experiment by 11.5% and 22.5% Root Mean Square Error (RMSE) gain in the whole and extreme domains of the label, respectively.

Optimization of Daily Trading Strategy

This study proposes a novel method that integrates variational pattern decomposition (VMD), long short-term memory (LSTM) models, and particle swarm optimization (PSO) to optimize daily trading strategies for a portfolio. By analyzing historical price data for gold and bitcoin, we have successfully identified the optimal investment actions (buy, hold, or sell) for a portfolio comprising different assets. The VMD-LSTM model proposed by us mitigates the long-term dependence problem of machine learning algorithm RNN in financial time series prediction through batch normalization optimization, and realizes the accurate prediction of the future price of gold and bitcoin, whose R-squared values reach 0.99 and 0.98 respectively. We also applied the particle swarm optimization algorithm to a sequence of trading decisions generated by predicting prices, ultimately obtaining a maximum portfolio value of $20,560,265.5, which is more than 14.2 times the 10-year growth of Bitcoin and 12,310.5 times the 10-year growth of gold. By verifying the universality and local optimal properties of our model, we use the Sharpe ratio as an objective function to evaluate the effect of the portfolio. After several perturbation tests, the results show that the original scheme is superior to other perturbation schemes considering the Sharpe ratio and the final return, which verifies the validity of the investment scheme selected by us as the local optimal value. These findings make our research more coherent and provide investors with a reliable basis for making decisions to achieve long-term capital appreciation and risk control.

Financial time series prediction under Covid-19 pandemic crisis with Long Short-Term Memory (LSTM) network

In this paper, we design and apply the Long Short-Term Memory (LSTM) neural network approach to predict several financial classes’ time series under COVID-19 pandemic crisis period. We use the S&P GSCI commodity indices and their sub-indices and consider the stock market indices for different regions. Based on the daily prices, the results show that the proposed LSTM network can form a robust prediction model to determine the optimal diversification strategies. Our prediction model achieved RMSEs and MAEs too small for the different selected financial assets, showing the predictive power of our LSTM network especially during the COVID-19 health crisis. In addition, our LSTM network outperforms ARIMA-type models for all selected assets.

Practical machine learning: Forecasting daily financial markets directions

Financial time series prediction has many applications in economics, but producing profitable strategies certainly has a special place among them, a daunting challenge. Statistical and machine learning techniques are intensively researched in the search for a holy grail of stock markets forecasting. However, it is not clear to prospecting researchers how good those popular models are regarding useful predictions on a real scenario. This paper contributes to that discussion, providing decisive evidences contrary to the use of basic out-of-the-box models, specifically Artificial Neural Networks (ANN), Support Vector Machines (SVM), Random Forest (RF) and Naive-Bayes (NB). Results consider optimistic and unreal variables often found in literature, as well as a more close-to-real simulation of the models usage. Specifically, current day closing prices direction forecasting results are contrasted with those on next day forecasts. As expected, when forecasting the current day, accuracy is almost perfect. However, when used to forecast next day closing direction, with a strict data separation policy and without direction or snooping bias, ANN, SVM, RF and NB produce results essentially equal to random guessing. The main achieved result is the demonstration of how a machine learning approach would fare in a support decision system for forecasting short-term future market direction, regardless of the level of market development, considering more than 100 securities in a 10 years period. Consequences for algorithmic trading relate to discouraging usage of the considered models as implemented here. On a more abstract sense, this paper presents more evidence to the Efficient Market Hypothesis (EMH).

Financial time series prediction using deep computing approaches

<p class="Abstracttitle">A financial time series is chaotic and non-stationary in nature, and predicting it outcomes is a very complex and challenging task. In this research, the theory of chaos, Long Short-Term Memory (LSTM), and Polynomial Regression (PR) are used in tandem to create a novel financial time series prediction hybrid, Chaos+LSTM+PR. The first step in this hybrid will determine whether or not a financial time series contains chaos. Following that, the chaos in the time series is modeled using Chaos Theory. The modeled time series is fed into the LSTM to obtain initial predictions. The error series obtained from LSTM predictions is fitted by PR to obtain error predictions. The error predictions and initial predictions from LSTM are combined to obtain final predictions. The effectiveness of this hybrid is examined by three types of financial time series (Chaos+LSTM+PR), including stock market indices (S&P 500, Nifty 50, Shanghai Composite), commodity prices (gold, crude oil, soya beans), and foreign exchange rates (INR/USD, JPY/USD, SGD/USD). The results show that the proposed hybrid outperforms ARIMA (autoregressive integrated moving average), Prophet, CART (Classification and Regression Tree), RF (Random Forest), LSTM, Chaos+CART, Chaos+CART, and Chaos+LSTM. The results are also checked for statistical significance.</p>

A Financial Time-Series Prediction Model Based on Multiplex Attention and Linear Transformer Structure

Financial time-series prediction has been an important topic in deep learning, and the prediction of financial time series is of great importance to investors, commercial banks and regulators. This paper proposes a model based on multiplexed attention mechanisms and linear transformers to predict financial time series. The linear transformer model has a faster model training efficiency and a long-time forecasting capability. Using a linear transformer reduces the original transformer’s complexity and preserves the decoder’s multiplexed attention mechanism. The results show that the proposed method can effectively improve the prediction accuracy of the model, increase the inference speed of the model and reduce the number of operations, which has new implications for the prediction of financial time series.

A Hybrid Model of Primary Ensemble Empirical Mode Decomposition and Quantum Neural Network in Financial Time Series Prediction

Financial time series are nonlinear, volatile and chaotic. Inspired by quantum computing, this paper proposed a new model, called primary ensemble empirical mode decomposition combined with quantum neural network (PEEMD-QNN) in predicting the stock index. PEEMD-QNN takes the advantages of the PEEMD which retains the main component of modal component and QNN. To demonstrate that our PEEMD-QNN model is robust, we used the new model to predict six major stock index time series in China at a specific time. Detailed experiments are implemented for both of the proposed prediction models, in which empirical mode decomposition combined with QNN (EMD-QNN), QNN and BP neural network are compared. The results demonstrate that the proposed PEEMD-QNN model has higher accuracy than BP neural network, QNN model and EMD-QNN model in stock market prediction.

An Efficient GAN-Based Multi-classification Approach for Financial Time Series Volatility Trend Prediction

Deep learning has achieved tremendous success in various applications owing to its robust feature representations of complex high-dimensional nonlinear data. Financial time-series prediction is no exception. Hence, the volatility trend prediction in financial time series (FTS) has been an active topic for several decades. Inspired by generative adversarial networks (GAN), which have been studied extensively in image processing and have achieved excellent results, we present the ordinal regression GAN for financial volatility trends (ORGAN-FVT) method for the end-to-end multi-classification task of FTS. An improved generative model based on convolutional long short-term memory (ConvLSTM) and multilayer perceptron (MLP) is proposed to capture temporal features effectively and mine the data distribution of volatility trends (short, neutral, and long) from given FTS data. Meanwhile, ordinal regression is leveraged for the discriminator to improve the multi-classification performance, making the model more practical. Finally, we empirically compare ORGAN-FVT with several state-of-the-art approaches on three real-world stock datasets: MICROSOFT(MSFT), Tesla(TSLA), and The People’s Insurance Company of China(PAICC). ORGAN-FVT demonstrated significantly better AUC and F1 scores, at most 20.81% higher than its competitors.

A Time Series Data Prediction Model Based on Adaptive Weighted LSTM

Financial time series prediction has always been a hot topic in the field of statistics learning. Aiming at the step selection problem of LSTM time series prediction model, this paper proposes an adaptive weighted LSTM model based on model average method. The model average is mainly reflected in two aspects: On the one hand, the proposed method takes intraday price information into account. Firstly, functional and nonlinear information of intraday price series are extracted through functional principal component analysis and kernel principal component analysis, and then Bagging is used to fit the residual sequence generated by the original LSTM model. On the other hand, the proposed method integrates the information of the model under different time Windows by using the weight based on distance correlation coefficient, and adaptively solves the step size selection problem, so as to improve the effectiveness of the overall model. The actual data analysis results show that the proposed method can effectively improve the prediction accuracy of the original LSTM model and has a certain robustness. Due to the flexibility of the proposed method, it can be used in time series prediction such as energy consumption prediction, environment detection and road traffic flow monitoring.