arXiv:2511.17275v1 Announce Type: new 
Abstract: Premium automotive manufacturers face increasingly complex forecasting challenges due to high product variety, sparse variant-level data, and volatile market dynamics. This study addresses monthly automobile demand forecasting across a multi-product, multi-market, and multi-level hierarchy using data from a German premium manufacturer. The methodology combines point and probabilistic forecasts across strategic and operational planning levels, leveraging ensembles of LightGBM models with pooled training sets, quantile regression, and a mixed-integer linear programming reconciliation approach. Results highlight that spatiotemporal dependencies, as well as rounding bias, significantly affect forecast accuracy, underscoring the importance of integer forecasts for operational feasibility. Shapley analysis shows that short-term demand is reactive, shaped by life cycle maturity, autoregressive momentum, and operational signals, whereas medium-term demand reflects anticipatory drivers such as online engagement, planning targets, and competitive indicators, with online behavioral data considerably improving accuracy at disaggregated levels.

تم إجراء دراسة حول توقعات الطلب على السيارات، مع التركيز على مصنع سيارات فاخرة ألماني. تستخدم الدراسة مجموعة من تقنيات النمذجة الزمانية والمكانية والهرمية، مستفيدة من نماذج LightGBM لتعزيز دقة التوقعات في ظل ديناميكيات سوق معقدة وتنوع كبير في المنتجات.

Se ha realizado un estudio sobre la previsión de la demanda de automóviles, centrándose en un fabricante premium alemán. La investigación emplea una combinación de técnicas de modelado espaciotemporal y jerárquico, utilizando modelos LightGBM para mejorar la precisión de las previsiones en medio de dinámicas de mercado complejas y una gran variedad de productos.

Une étude a été réalisée sur la prévision de la demande automobile, en se concentrant sur un fabricant premium allemand. La recherche utilise une combinaison de techniques de modélisation spatiotemporelle et hiérarchique, en s'appuyant sur des modèles LightGBM pour améliorer la précision des prévisions face à des dynamiques de marché complexes et à une grande variété de produits.

A study has been conducted on automobile demand forecasting, focusing on a German premium manufacturer. The research employs a combination of spatiotemporal and hierarchical modeling techniques, utilizing LightGBM models to enhance forecast accuracy amidst complex market dynamics and product variety.

Automobile demand forecasting: Spatiotemporal and hierarchical modeling, life cycle dynamics, and user-generated online information

arXiv:2511.19090v1 Announce Type: new 
Abstract: The advent of financial technology has witnessed a surge in the utilization of deep learning models to anticipate consumer conduct, a trend that has demonstrated considerable potential in enhancing lending strategies and bolstering market efficiency. We study multi-horizon demand forecasting on e-commerce transactions using the UCI Online Retail II dataset. Unlike prior versions of this manuscript that mixed financial-loan narratives with retail data, we focus exclusively on retail market behavior and define a clear prediction target: per SKU daily demand (or revenue) for horizons H=1,7,14. We present a hybrid sequence model that combines multi-scale temporal convolutions, a gated recurrent module, and time-aware self-attention. The model is trained with standard regression losses and evaluated under MAE, RMSE, sMAPE, MASE, and Theil's U_2 with strict time-based splits to prevent leakage. We benchmark against ARIMA/Prophet, LSTM/GRU, LightGBM, and state-of-the-art Transformer forecasters (TFT, Informer, Autoformer, N-BEATS). Results show consistent accuracy gains and improved robustness on peak/holiday periods. We further provide ablations and statistical significance tests to ensure the reliability of improvements, and we release implementation details to facilitate reproducibility.

تبرز الدراسة الأخيرة حول تحسين نماذج التعلم العميق لتوقع سلوك السوق الديناميكي فعالية نموذج تسلسلي هجين في توقع معاملات التجارة الإلكترونية. يستخدم هذا النموذج التلافيف الزمنية متعددة المقاييس، ووحدات متكررة ذات بوابات، والانتباه الذاتي الزمني لتوقع الطلب اليومي بدقة على SKU فردية عبر عدة آفاق. تستند الأبحاث إلى مجموعة بيانات UCI Online Retail II وتستخدم مقاييس تقييم صارمة لضمان الموثوقية.

El reciente estudio sobre la optimización de modelos de aprendizaje profundo para predecir el comportamiento dinámico del mercado destaca la efectividad de un modelo secuencial híbrido en la previsión de transacciones de comercio electrónico. Este modelo utiliza convoluciones temporales de múltiples escalas, módulos recurrentes con puertas y atención auto-consciente para predecir con precisión la demanda diaria de SKU individuales en múltiples horizontes. La investigación se basa en el conjunto de datos UCI Online Retail II y emplea métricas de evaluación rigurosas para garantizar la fiabilida…

La récente étude sur l'optimisation des modèles d'apprentissage profond pour prédire le comportement dynamique du marché met en évidence l'efficacité d'un modèle séquentiel hybride dans la prévision des transactions de commerce électronique. Ce modèle utilise des convolutions temporelles multi-échelles, des modules récurrents à portes et une attention auto-consciente pour prédire avec précision la demande quotidienne pour des SKU individuels sur plusieurs horizons. La recherche est basée sur le jeu de données UCI Online Retail II et utilise des métriques d'évaluation rigoureuses pour garantir …

The recent study on the optimization of deep learning models for predicting dynamic market behavior highlights the effectiveness of a hybrid sequence model in forecasting e-commerce transactions. This model utilizes multi-scale temporal convolutions, gated recurrent modules, and time-aware self-attention to accurately predict daily demand for individual SKUs over multiple horizons. The research is based on the UCI Online Retail II dataset and employs rigorous evaluation metrics to ensure reliability.

Optimization of Deep Learning Models for Dynamic Market Behavior Prediction

arXiv:2506.08274v5 Announce Type: replace-cross 
Abstract: This research addresses the critical lack of comprehensive studies on feature scaling by systematically evaluating 12 scaling techniques - including several less common transformations - across 14 different Machine Learning algorithms and 16 datasets for classification and regression tasks. We meticulously analyzed impacts on predictive performance (using metrics such as accuracy, MAE, MSE, and $R^2$) and computational costs (training time, inference time, and memory usage). Key findings reveal that while ensemble methods (such as Random Forest and gradient boosting models like XGBoost, CatBoost and LightGBM) demonstrate robust performance largely independent of scaling, other widely used models such as Logistic Regression, SVMs, TabNet, and MLPs show significant performance variations highly dependent on the chosen scaler. This extensive empirical analysis, with all source code, experimental results, and model parameters made publicly available to ensure complete transparency and reproducibility, offers model-specific crucial guidance to practitioners on the need for an optimal selection of feature scaling techniques.

دراسة حديثة نُشرت على arXiv قامت بتقييم منهجي لـ 12 تقنية لتقييس الميزات عبر 14 خوارزمية تعلم آلي و16 مجموعة بيانات، كاشفة عن تباينات كبيرة في الأداء في نماذج مثل الانحدار اللوجستي وSVM، بينما أظهرت طرق التجميع قوة مستقلة عن التقييس.

Un estudio reciente publicado en arXiv evaluó sistemáticamente 12 técnicas de escalado de características a través de 14 algoritmos de aprendizaje automático y 16 conjuntos de datos, revelando variaciones significativas en el rendimiento de modelos como la regresión logística y los SVM, mientras que los métodos de ensamblaje mostraron robustez independientemente del escalado.

Une étude récente publiée sur arXiv a évalué systématiquement 12 techniques de mise à l'échelle des caractéristiques à travers 14 algorithmes d'apprentissage automatique et 16 ensembles de données, révélant des variations de performance significatives dans des modèles tels que la régression logistique et les SVM, tandis que les méthodes d'ensemble ont montré une robustesse indépendamment de la mise à l'échelle.

A recent study published on arXiv systematically evaluated 12 feature scaling techniques across 14 machine learning algorithms and 16 datasets, revealing significant performance variations in models like Logistic Regression and SVMs, while ensemble methods showed robustness regardless of scaling.

Automobile demand forecasting: Spatiotemporal and hierarchical modeling, life cycle dynamics, and user-generated online information

Was this article worth reading? Share it

Gestell

CarsXE

MicroEstimates