Leakage-Controlled Horizon-Specific Model Selection for Daily Equity Forecasting: An Automated Multi-Model Pipeline

Accurate forecasting of financial time series increasingly relies on alternative data such as environmental, social and governance (ESG) scores and news-based sentiment, yet the way these signals interact and when they actually improve forecasts is still poorly understood. We introduce an interpretable hybrid framework for asset return forecasting that combines a Temporal Fusion Transformer (TFT) with a lightweight Support Vector Regression (SVR) residual corrector and an explicit gated late fusion of ESG features with aspect-based financial sentiment (FinBERT-based ABSA). The gating mechanism learns when to emphasize sustainability versus sentiment signals, while SHAP interaction values and Friedman's H quantify ESG-sentiment interactions across assets and regimes. A finance-grade, leak-proof walk-forward protocol (252 trading days train / 10 days test, within-fold scaling, ABSA items strictly before 16:00ET; ESG effective T+3; macro T+1, HAC-robust Diebold-Mariano tests) is applied to US large-cap technology equities, major global indices, and BTC/ETH over 2020-2024. Across [Formula: see text] independent seeds, the hybrid achieves aggregate mean absolute error of [Formula: see text] and RMSE of [Formula: see text] on next-day log returns, with directional accuracy [Formula: see text], IC 0.39, and ICIR 0.82, significantly outperforming tuned deep-learning and machine-learning baselines (HAC-robust per-asset Diebold-Mariano tests with BH-FDR [Formula: see text]; Fisher aggregation yields [Formula: see text]). Simple long-only, thresholded simulations indicate higher risk-adjusted performance and lower maximum drawdown under conservative transaction-cost assumptions. Ablation studies show that removing either ESG or sentiment features yields the largest degradations, and that the SVR corrector stabilizes errors under regime shifts. To directly address market-cycle sensitivity, we evaluate stability across event-defined stress windows (COVID-19 crash, 2022 tightening cycle, and 2023 banking stress) and volatility-defined regimes using terciles of 20-day realized volatility. We report regime-split forecasting and strategy metrics with block-bootstrap confidence intervals, HAC-robust Diebold-Mariano tests within each regime, and residual-stabilization diagnostics that quantify the SVR variance and skewness reduction under stress. ESG-sentiment interactions are statistically non-zero and regime-dependent, with sentiment gaining importance in turbulent periods and ESG in calmer markets. A latency-optimized variant that removes auxiliary BiLSTMs retains over [Formula: see text] of the accuracy gains while reducing inference time by approximately [Formula: see text] of the full model (i.e., a reduction of about [Formula: see text]), supporting near-real-time deployment.

A novel type of learning machine called support vector machine (SVM) has been receiving increasing interest in areas ranging from its original application in pattern recognition to other applications such as regression estimation due to its remarkable generalization performance. This paper deals with the application of SVM in financial time series forecasting. The feasibility of applying SVM in financial forecasting is first examined by comparing it with the multilayer back-propagation (BP) neural network and the regularized radial basis function (RBF) neural network. The variability in performance of SVM with respect to the free parameters is investigated experimentally. Adaptive parameters are then proposed by incorporating the nonstationarity of financial time series into SVM. Five real futures contracts collated from the Chicago Mercantile Market are used as the data sets. The simulation shows that among the three methods, SVM outperforms the BP neural network in financial forecasting, and there are comparable generalization performance between SVM and the regularized RBF neural network. Furthermore, the free parameters of SVM have a great effect on the generalization performance. SVM with adaptive parameters can both achieve higher generalization performance and use fewer support vectors than the standard SVM in financial forecasting.

Financial time series forecasting with deep learning : A systematic literature review: 2005–2019

Financial time series forecasting plays a crucial role in predicting future market trends, pricing assets, and managing risks in financial markets. This paper compares traditional methods, such as ARIMA, Exponential Smoothing, and GARCH, with AI-driven techniques, including machine learning and deep learning models, for financial time series forecasting. Traditional models are well-established and effective for stationary data, but they struggle with non-linear relationships and large datasets. In contrast, AI-driven techniques, such as Random Forests, Long Short-Term Memory Networks (LSTMs), and reinforcement learning, offer improved accuracy and adaptability by capturing complex patterns in the data. However, these models come with higher computational complexity and challenges related to interpretability. The paper provides a comprehensive comparison of these methods, highlighting their strengths, limitations, and practical applications. It concludes by offering recommendations for when to use traditional methods versus AI-driven approaches, based on the nature of the data and forecasting needs. The integration of AI with traditional models is also discussed as a promising future direction in financial forecasting.

This literature review provides a comprehensive overview of key developments in quantitative trading theory, machine learning-based financial time series forecasting, deep learning-based financial time series forecasting, and modern quantitative investment strategies. It highlights seminal contributions from renowned scholars and researchers in the field. This review first explores the Efficient Market Hypothesis (EMH) proposed by Eugene Fama in 1970 and its empirical validation, which lays the foundation for understanding stock market dynamics. It then focuses on the application of machine learning to financial time series forecasting, including Hulls Delta strategy, Hujll Js multifactor model regression series, and Junhua Chens seminal work in 2009, which emphasizes the role of machine learning in solving the challenges of financial time series data. Finally, an overview of the development of deep learning in financial time series forecasting is presented, including the comprehensive model of Bowie et al., the success of Junhua Chens Deep Belief Network (DBN), and the sequence data processing capabilities of the Multi-stage Attention Network (MAN) model. In terms of modern quantitative investment strategies, it covers research areas such as EBIT/EV-based stock ranking studies, higher-order moment analysis, frequent trading strategies, and investor sentiment indicators. In summary, this literature review showcases the evolution of quantitative trading theory, the emergence of machine learning and deep learning in financial time series forecasting, and the development of modern quantitative investment strategies, offering valuable insights and tools for investors, researchers, and practitioners in financial markets.

As the complexity of financial markets continues to increase, traditional financial time series forecasting methods face huge challenges. Machine learning, especially deep learning, has become a powerful tool to meet this challenge. However, these models often require a lot of computing resources when processing large-scale data, resulting in delays in the training and prediction process. In order to improve the efficiency and real-time performance of prediction, machine learning algorithm acceleration technology has emerged, mainly through hardware acceleration and software optimization to improve the speed of model training and reasoning. This study explores the application of machine learning algorithm acceleration technology in financial time series forecasting, analyzes how acceleration technology can help solve problems in data processing, real-time forecasting, and other aspects, and demonstrates the actual effects of these technologies in the stock market, foreign exchange market, and commodity market through case analysis. Despite the challenges of data diversity and hardware resource limitations, future research directions will focus on the combination of deep reinforcement learning, cloud computing, edge computing, and quantum computing to further promote the intelligent and efficient development of financial time series forecasting.

This study proposes a novel hybrid forecasting approach designed explicitly for long-horizon financial time series. It incorporates LMD (Local Mean Decomposition), SD (Signal Decomposition), and sophisticated machine learning methods. The framework for the NASDAQ Composite Index begins by decomposing the original time series into stochastic and deterministic components using the LMD approach. This method effectively separates linear and nonlinear signal structures. The stochastic components are modeled using ARIMA to represent linear temporal dynamics, while the deterministic components are projected using cutting-edge machine learning methods, including XGBoost, Random Forest (RF), Artificial Neural Networks (ANNs), and Support Vector Machines (SVMs). This study employs various statistical metrics to evaluate the predictive ability across both short-term noise and long-term trends, including Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), and Directional Statistic (DS). Furthermore, the Diebold–Mariano test is used to determine the statistical significance of any forecast improvements. Empirical results demonstrate that the hybrid LMD–ARIMA–SD–XGBoost model consistently outperforms alternative configurations in terms of prediction accuracy and directional consistency. These findings demonstrate the advantages of integrating decomposition-based signal filtering with ensemble machine learning to improve the robustness and generalizability of long-term forecasting. This study presents a scalable and adaptive approach for modeling complex, nonlinear, and high-dimensional time series, thereby contributing to the enhancement of intelligent forecasting systems in the economic and financial sectors. As far as the authors are aware, this is the first study to combine XGBoost and LMD in a hybrid decomposition framework for forecasting long-horizon stock indexes.

Financial forecasting is indispensable for various financial sectors' strategic decision-making, risk management, and investment planning. Traditional statistical methods, though valuable, frequently struggle to capture the intricate, non-linear patterns embedded within financial data. In response to these limitations, Gradient Boosted Trees (GBTs) have emerged as a formidable ensemble learning technique renowned for enhancing predictive accuracy. This comprehensive review paper delves into applying GBTs in financial forecasting, scrutinizing their methodological underpinnings, distinct advantages, and comparative performance against other machine learning approaches. GBTs, with their real-world applications in financial forecasting, operate by sequentially combining multiple weak learners, typically decision trees, to iteratively refine predictions by focusing on residual errors. This iterative approach enables GBTs to effectively model complex relationships and interactions in financial datasets, outperforming traditional models like ARIMA and linear regression in many scenarios. Moreover, the paper addresses critical implementation challenges associated with GBTs, such as hyperparameter tuning and computational complexity, which are pivotal for achieving optimal performance. The review identifies promising avenues for future research, including integrating GBTs with deep learning techniques and advancements in real-time forecasting capabilities. By elucidating these aspects, this paper aims to provide insights that enhance the application and efficacy of GBTs in financial forecasting contexts.

This research evaluates the predictive performance and uncertainty quantification of traditional and Bayesian approaches at ARIMA models applied to the S&P 500 dataset. Both models were assessed using mean absolute error (MAE) and root mean squared error (RMSE). The Bayesian ARIMA model demonstrated better predictive accuracy, with lower MAE (467.79) and RMSE (467.86) values compared to the traditional ARIMA model (MAE: 473.75, RMSE: 473.68). The probabilistic structure of the Bayesian ARIMA model provides an advantage by effectively managing stock price volatility and uncertainty. Unlike traditional ARIMA, which offers point estimates, Bayesian ARIMA generates a posterior distribution of forecasts that captures model uncertainty and includes credible intervals, offering deeper insights into future predictions. This feature is particularly valuable in financial forecasting, where uncertainty plays a key role. The findings indicate that Bayesian ARIMA is more robust and reliable for forecasting in uncertain and volatile economic environments. For this analysis, the forecast and bayesforecast packages in R were used, enabling efficient implementation of the models. Consequently, Bayesian ARIMA is recommended over traditional ARIMA for financial time series forecasting in highly volatile markets.. KEYWORDS :Bayesian, ARIMA, Predictive accuracy, Uncertainty quantification, S&P 500, Financial forecasting, MAE, RMSE, Posterior distribution, Credible intervals.

Application of support vector machines in financial time series forecasting

The inherent challenges of financial time series forecasting demand advanced modeling techniques for reliable predictions. Effective financial time series forecasting is crucial for financial risk management and the formulation of investment decisions. The accurate prediction of stock prices is a subject of study in the domains of investing and national policy. This problem appears to be challenging due to the presence of multi-noise, nonlinearity, volatility, and the chaotic nature of stocks. This paper proposes a novel financial time series forecasting model based on the deep learning ensemble model LSTM-mTrans-MLP, which integrates the long short-term memory (LSTM) network, a modified Transformer network, and a multilayered perception (MLP). By integrating LSTM, the modified Transformer, and the MLP, the suggested model demonstrates exceptional performance in terms of forecasting capabilities, robustness, and enhanced sensitivity. Extensive experiments are conducted on multiple financial datasets, such as Bitcoin, the Shanghai Composite Index, China Unicom, CSI 300, Google, and the Amazon Stock Market. The experimental results verify the effectiveness and robustness of the proposed LSTM-mTrans-MLP network model compared with the benchmark and SOTA models, providing important inferences for investors and decision-makers.

Financial time series forecasting using independent component analysis and support vector regression

This article proposes a lightweight deep learning model based on residual thinking, RAR-LSTM (Residual and Regime aware LSTM), to address the difficult issues of high noise, non stationarity, and state switching in economic and financial time series forecasting. This method uses interpretable naive models as baselines (such as seasonal naive method for quarterly forecasting and random walk method for financial daily frequency forecasting), and only learns the residual part through small-scale LSTM, thus balancing stability and reproducibility. At the same time, introducing minimalist state features (such as quarterly dummy variables, high and low rolling volatility states) helps the model perceive changes in the macro or market environment; In terms of loss function, Pinball Loss is supported for quantile prediction to meet the demand for asymmetric underestimation and overestimation costs in practical business. This article validates the generality of RAR-LSTM in two typical tasks: the macro quarterly task is to predict the US unemployment rate, and the financial daily frequency task is to predict the volatility of S&P 500 days. The experimental results show that RAR-LSTM significantly outperforms methods such as naive baseline, traditional LSTM, and SARIMA in terms of prediction accuracy (MAE, RMSE, etc.), especially in high volatility states where it performs more robustly. This model has the characteristics of simple structure, low computational cost, and easy reproducibility, and is suitable for economic and financial forecasting tasks in teaching, competition, and practical scenarios. Beyond naive baselines, we benchmark against stronger statistical models (ARIMA and ETS) and show that RAR-LSTM remains robust across baselines. We also report quantile forecasts (τ=0.1/0.5/0.9) using Pinball Loss, demonstrating calibrated coverage and practical benefits for risk-aware decisions.

ABSTRACTIntuitionistic fuzzy sets (IFSs) have been proved to be more ideal than fuzzy sets to handle non-probabilistic uncertainty and non-determinism in the system. The present study proposes an IFS-based computational method to address the issue of non-determinism in financial time series forecasting. The proposed IFS-based forecasting method uses a simple computational algorithm to forecast without using complex computations using intuitionistic fuzzy logical relations. In order to see suitability of the proposed method in financial forecasting, it is implemented on three experimental data of the SBI share price, TAIEX and Dow Jones Industrial Average (DJIA). Root mean square error and statistical test are used in the study to confirm the out performance of the proposed IFS-based computational method of forecasting. Experimental results show that the proposed method outperforms various existing methods for forecasting SBI, TAIEX and DJIA.

This paper is discusses the problems of the short-term forecasting of financial time series using supervised machine learning (ML) approach. For this goal, we applied several the most powerful methods including Support Vector Machine (SVM), Multilayer Perceptron (MLP), Random Forests (RF) and Stochastic Gradient Boosting Machine (SGBM). As dataset were selected the daily close prices of two stock index: SP 500 and NASDAQ, two the most capitalized cryptocurrencies: Bitcoin (BTC), Ethereum (ETH), and exchange rate of EUR-USD. As features we used only the past price information. To check the efficiency of these models we made out-of-sample forecast for selected time series by using one step ahead technique. The accuracy rates of the forecasted prices by using ML models were calculated. The results verify the applicability of the ML approach for the forecasting of financial time series. The best out of sample accuracy of short-term prediction daily close prices for selected time series obtained by SGBM and MLP in terms of Mean Absolute Percentage Error (MAPE) was within 0.46-3.71 %. Our results are comparable with accuracy obtained by Deep learning approaches.