Journal of Forecasting最新文献_第4页

TiF: A Multi-Scale Data Fusion and Fourier Encoding Framework for Financial Risk Prediction 金融风险预测的多尺度数据融合与傅立叶编码框架

IF 2.6 3区经济学

Journal of Forecasting Pub Date : 2026-08-05 Epub Date: 2026-05-12 DOI: 10.1002/for.70172

Haomin Zhang, Puyu Zhou

{"title":"TiF: A Multi-Scale Data Fusion and Fourier Encoding Framework for Financial Risk Prediction","authors":"Haomin Zhang, Puyu Zhou","doi":"10.1002/for.70172","DOIUrl":"https://doi.org/10.1002/for.70172","url":null,"abstract":"<div>\u0000 \u0000 <p>Financial forecasting often treats annual, quarterly, and monthly information in isolation, which harms robustness in volatile markets. We present <i>Time-integrated Fourier (TiF)</i>, a lightweight framework that fuses annual, quarterly, and monthly signals via energy-based Fourier encoding and a compact 1D-CNN backbone with learnable cross-timescale gating. TiF is evaluated under a standardized rolling-origin expanding-window protocol on a core technology benchmark (Apple, Microsoft, Adobe) and further examined on two external validations: (A) a non-financial EBITDA cohort (\u0000<span></span><math>\u0000 <mi>N</mi>\u0000 <mo>=</mo>\u0000 <mn>105</mn></math>) and (B) a finance-sector cohort with ROAA as the target. Across studies, some baselines attain the best single-metric scores (e.g., a Transformer for \u0000<span></span><math>\u0000 <msup>\u0000 <mrow>\u0000 <mi>R</mi>\u0000 </mrow>\u0000 <mrow>\u0000 <mn>2</mn>\u0000 </mrow>\u0000 </msup></math>, CatBoost for RMSE), whereas TiF delivers consistently strong and <i>balanced</i> performance over \u0000<span></span><math>\u0000 <msup>\u0000 <mrow>\u0000 <mi>R</mi>\u0000 </mrow>\u0000 <mrow>\u0000 <mn>2</mn>\u0000 </mrow>\u0000 </msup></math>/RMSE/MAPE with small variance and modest parameter counts. Ablations quantify the contribution of Fourier bands and each timescale branch, and robustness checks confirm the same qualitative ranking across winsorization/clipping settings. All experiments share identical preprocessing, validation length, and hyperparameter budgets, and we report mean \u0000<span></span><math>\u0000 <mo>±</mo></math> SD across rolling origins and seeds to ensure fair comparison.</p>\u0000 </div>","PeriodicalId":47835,"journal":{"name":"Journal of Forecasting","volume":"45 6","pages":"3030-3040"},"PeriodicalIF":2.6,"publicationDate":"2026-08-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148696524","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Forecasting With Dynamic Factor Models Estimated by Partial Least Squares 用偏最小二乘估计的动态因子模型进行预测

IF 2.6 3区经济学

Journal of Forecasting Pub Date : 2026-08-05 Epub Date: 2026-04-17 DOI: 10.1002/for.70158

Samuel Rauhala

引用次数: 0

SGAT-SP: Sparse Graph Attention Network for Stock Prediction 股票预测的稀疏图关注网络

IF 2.6 3区经济学

Journal of Forecasting Pub Date : 2026-08-05 Epub Date: 2026-05-03 DOI: 10.1002/for.70169

Gautam Gupta, Priyanshi Goel, Divyanshi Verma, Amarjit Malhotra

{"title":"SGAT-SP: Sparse Graph Attention Network for Stock Prediction","authors":"Gautam Gupta, Priyanshi Goel, Divyanshi Verma, Amarjit Malhotra","doi":"10.1002/for.70169","DOIUrl":"https://doi.org/10.1002/for.70169","url":null,"abstract":"<div>\u0000 \u0000 <p>Industry experts as well as academic scholars have directed substantial attention toward researching stock market volatility. Most of the past research has focused on using singular features such as closing prices, opening prices, or news stories to predict stock movement. In recent times, there has been a growing interest in using multimodal graph neural networks which can analyze a variety of features. Most of these methods focus on node aggregation to extract relevant features from related stocks to further improve model accuracy. The current state-of-the-art model—ML-GAT: Multilevel Graph Attention Model—constructs a graph network between the stocks using Wikidata relations. It uses multiple layers of graph attention to aggregate features such as historical price features and current news. However, the high number of intercompany relationships in ML-GAT may include irrelevant and noisy edges. It also uses individual attention coefficients for each layer, leading to inefficient utilization of computational resources. To overcome these challenges, a Sparse Graph Attention Network for Stock Prediction (SGAT-SP) is proposed in this paper. SGAT-SP uses a single set of attention coefficients to reduce training and inference time. It assigns a binary mask to every edge which represents whether they will be used for node aggregation to reduce noisy edges. The proposed approach achieves an average accuracy score of 0.83, a slight improvement over ML-GAT, which has an accuracy score of 0.827. Additionally, it significantly reduces inference time by 85%, resulting in faster results and decreased computational expenses.</p>\u0000 </div>","PeriodicalId":47835,"journal":{"name":"Journal of Forecasting","volume":"45 6","pages":"2942-2953"},"PeriodicalIF":2.6,"publicationDate":"2026-08-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148696092","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Selecting Forecasting Methods via Literature Prominence and Accuracy Testing: Evidence From Container Throughput Forecasting 通过文献显著性和准确性检验选择预测方法：来自集装箱吞吐量预测的证据

IF 2.6 3区经济学

Journal of Forecasting Pub Date : 2026-08-05 Epub Date: 2026-05-13 DOI: 10.1002/for.70178

Kamal Sanguri

{"title":"Selecting Forecasting Methods via Literature Prominence and Accuracy Testing: Evidence From Container Throughput Forecasting","authors":"Kamal Sanguri","doi":"10.1002/for.70178","DOIUrl":"https://doi.org/10.1002/for.70178","url":null,"abstract":"<div>\u0000 \u0000 <p>Selecting methods for monthly container throughput forecasting has become more difficult in the post-COVID period because the literature now offers a wider set of classical, decomposition-based, machine learning, and deep learning approaches. This study develops a literature-based Preference Index (PI) to capture method-family prominence and combines it with rolling-origin forecast evaluation using monthly throughput data, measured in twenty-foot equivalent units (TEUs), from the Port of Los Angeles, the Port of Singapore Authority, and Hong Kong Port. The PI combines use, citation influence, and co-mention embeddedness to shortlist method families rather than to rank them by forecasting superiority. The empirical phase evaluates representative univariate models: Error–Trend–Seasonal models (ETS), seasonal autoregressive integrated moving average models (SARIMA), seasonal-trend decomposition using Loess (STL)-based models as representatives of the Ensemble/Hybrid PI family, artificial neural network/multilayer perceptron models (ANN/MLP), and long short-term memory/recurrent neural network models (LSTM/RNN), across 1-, 3-, 6-, and 12-month horizons. Accuracy is assessed using mean absolute scaled error (MASE), mean absolute percentage error (MAPE), root mean squared error (RMSE), Diebold–Mariano tests, and Model Confidence Set (MCS) analysis. Results show that classical and decomposition-based methods remain strong forecasting anchors, but the best specification differs by port: STL + ARIMA performs best for Los Angeles, STL + ETS and SARIMA are strongest for Singapore depending on horizon, while ETS and SARIMA lead for Hong Kong. The lightly specified univariate neural models do not consistently outperform these alternatives. The study offers a transparent screen-and-test workflow and helps clarify when classical models may suffice and when learning-based approaches may require richer information.</p>\u0000 </div>","PeriodicalId":47835,"journal":{"name":"Journal of Forecasting","volume":"45 6","pages":"3041-3057"},"PeriodicalIF":2.6,"publicationDate":"2026-08-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148696403","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Forecasting Financial Risk Using Quantile Random Forests 用分位数随机森林预测金融风险

IF 2.6 3区经济学

Journal of Forecasting Pub Date : 2026-07-03 Epub Date: 2026-03-16 DOI: 10.1002/for.70146

Robert James, Jessica Wai Yin Leung

引用次数: 0

Using DSGE and Machine Learning to Forecast Public Debt for France 使用DSGE和机器学习预测法国公共债务

IF 2.6 3区经济学

Journal of Forecasting Pub Date : 2026-07-03 Epub Date: 2026-03-05 DOI: 10.1002/for.70144

Emmanouil Sofianos, Thierry Betti, Theophilos Papadimitriou, Amélie Barbier-Gauchard, Periklis Gogas

{"title":"Using DSGE and Machine Learning to Forecast Public Debt for France","authors":"Emmanouil Sofianos, Thierry Betti, Theophilos Papadimitriou, Amélie Barbier-Gauchard, Periklis Gogas","doi":"10.1002/for.70144","DOIUrl":"https://doi.org/10.1002/for.70144","url":null,"abstract":"<p>Forecasting public debt is essential for effective policymaking and economic stability, yet traditional approaches face challenges due to data scarcity. While machine learning (ML) has demonstrated success in financial forecasting, its application to macroeconomic forecasting remains underexplored, hindered by short historical time series and low-frequency (e.g., quarterly/annual) data availability. This study proposes a novel hybrid framework integrating dynamic stochastic general equilibrium (DSGE) modeling with ML techniques to address these limitations, focusing on the evolution of France's public debt. We first generate a large artificial macroeconomic dataset using an estimated DSGE model for France, which allows for efficient training of ML algorithms. These trained models are then applied to actual historical data for directional debt forecasting. The results show that the best machine learning model is an XGBoost achieving 90% accuracy, outperforming an elastic net model, used as benchmark. Our results highlight the viability of combining structural economic models with data-driven techniques to improve macroeconomic forecasting.</p>","PeriodicalId":47835,"journal":{"name":"Journal of Forecasting","volume":"45 5","pages":"2173-2185"},"PeriodicalIF":2.6,"publicationDate":"2026-07-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/for.70144","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148386774","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Robust Prediction Intervals for Time Series Forecasting: A Bootstrap and Bayesian Approach 时间序列预测的鲁棒预测区间：自举和贝叶斯方法

IF 2.6 3区经济学

Journal of Forecasting Pub Date : 2026-07-03 Epub Date: 2026-03-12 DOI: 10.1002/for.70126

Betty X.Y. Chu, Cathy W.S. Chen

{"title":"Robust Prediction Intervals for Time Series Forecasting: A Bootstrap and Bayesian Approach","authors":"Betty X.Y. Chu, Cathy W.S. Chen","doi":"10.1002/for.70126","DOIUrl":"https://doi.org/10.1002/for.70126","url":null,"abstract":"<div>\u0000 \u0000 <p>This study proposes a new bootstrap strategy, the standardized skewed Student-<i>t</i> block multiplier bootstrap (skew-\u0000<span></span><math>\u0000 <mi>t</mi></math>BMM), designed to better capture heteroskedasticity and heavy-tailed behavior in time series forecasting with machine learning models. We construct prediction intervals by integrating bootstrap resampling with machine learning forecasting methods and Bayesian approaches, including random forest, support vector regression, XGBoost, and long short-term memory networks (LSTM), while using the Bayesian structural time series (BSTS) and autoregressive integrated moving average (ARIMA)-type models as probabilistic benchmark references. To facilitate fair comparisons across models and data-generating mechanisms, we introduce a scale-adjusted relative interval score for evaluating interval forecasts. Extensive simulation studies covering nonlinear dynamics, deterministic seasonality, transfer-function effects, and conditional heteroskedasticity show that the proposed skew-\u0000<span></span><math>\u0000 <mi>t</mi></math>BMM delivers prediction intervals with improved calibration and competitive sharpness. An empirical application to monthly suicide counts in Japan illustrates the practical relevance of the proposed framework. Overall, the results show that machine learning models combined with the skew-\u0000<span></span><math>\u0000 <mi>t</mi></math>BMM, particularly LSTM, deliver reliable and accurate forecast distributions in complex and heavy-tailed environments, while BSTS performs particularly well for probabilistic forecasting.</p>\u0000 </div>","PeriodicalId":47835,"journal":{"name":"Journal of Forecasting","volume":"45 5","pages":"2309-2331"},"PeriodicalIF":2.6,"publicationDate":"2026-07-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148386488","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Cross-Check of Economic Forecasts 经济预测的交叉核对

IF 2.6 3区经济学

Journal of Forecasting Pub Date : 2026-07-03 Epub Date: 2026-02-27 DOI: 10.1002/for.70140

Frida Bowe, Eleonora Granziera, Pål B. Ulvedal

引用次数: 0

Global Insights Into Term Spreads: Unveiling Their Predictive Power During Unconventional Monetary Policy 对期限息差的全球洞察：揭示其在非常规货币政策中的预测能力

IF 2.6 3区经济学

Journal of Forecasting Pub Date : 2026-07-03 Epub Date: 2026-03-05 DOI: 10.1002/for.70133

Petri Kuosmanen, Juuso Vataja

引用次数: 0

Exploring the Nexus Between Sustainability Index and Central European Stock Markets Competitiveness: Evidence Through Time–Frequency Analysis and SHAP 可持续发展指数与中欧股票市场竞争力的关系研究：基于时频分析和SHAP的证据

IF 2.6 3区经济学

Journal of Forecasting Pub Date : 2026-07-03 Epub Date: 2026-03-06 DOI: 10.1002/for.70141

Zuzana Janková, Dipak Kumar Jana, Michal Páleš, Elena Fleaca, Petr Dostál

{"title":"Exploring the Nexus Between Sustainability Index and Central European Stock Markets Competitiveness: Evidence Through Time–Frequency Analysis and SHAP","authors":"Zuzana Janková, Dipak Kumar Jana, Michal Páleš, Elena Fleaca, Petr Dostál","doi":"10.1002/for.70141","DOIUrl":"https://doi.org/10.1002/for.70141","url":null,"abstract":"<p>Sustainability has become an important factor shaping financial markets and investor behavior. This paper examines the relationship between sustainability indices and Central European stock markets using a time–frequency approach. Wavelet coherence is employed to capture time-varying co-movements between sustainability indices and stock market returns in the Czech Republic, Poland, Hungary, and Slovakia. In addition, SHapley Additive exPlanations (SHAP) are used as a descriptive tool to summarize contemporaneous associations between sustainability indices and stock market returns. The results reveal strong regime dependence. During the COVID-19 pandemic, sustainability indices exhibited pronounced coherence with stock markets across both short- and long-term frequencies, indicating heightened interdependence. In contrast, during the period associated with the war in Ukraine, the relationship is mainly confined to high-frequency dynamics. Substantial cross-market heterogeneity is observed, with stronger linkages for the Czech, Polish, and Hungarian markets and weaker or negative associations for the Slovak market. A complementary forecasting analysis shows that sustainability-related information provides only limited short-term predictive power for daily returns once standard market-wide factors are controlled for. Overall, the findings suggest that sustainability influences stock markets primarily through time-varying and regime-dependent channels rather than through consistent short-term return predictability.</p>","PeriodicalId":47835,"journal":{"name":"Journal of Forecasting","volume":"45 5","pages":"2213-2237"},"PeriodicalIF":2.6,"publicationDate":"2026-07-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/for.70141","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148386629","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0