arXiv:2512.02092v1 Announce Type: cross 
Abstract: Timely assessment of current conditions is essential especially for small, open economies such as Singapore, where external shocks transmit rapidly to domestic activity. We develop a real-time nowcasting framework for quarterly GDP growth using a high-dimensional panel of approximately 70 indicators, encompassing economic and financial indicators over 1990Q1-2023Q2. The analysis covers penalized regressions, dimensionality-reduction methods, ensemble learning algorithms, and neural architectures, benchmarked against a Random Walk, an AR(3), and a Dynamic Factor Model. The pipeline preserves temporal ordering through an expanding-window walk-forward design with Bayesian hyperparameter optimization, and uses moving block-bootstrap procedures both to construct prediction intervals and to obtain confidence bands for feature-importance measures. It adopts model-specific and XAI-based explainability tools. A Model Confidence Set procedure identifies statistically superior learners, which are then combined through simple, weighted, and exponentially weighted schemes; the resulting time-varying weights provide an interpretable representation of model contributions. Predictive ability is assessed via Giacomini-White tests. Empirical results show that penalized regressions, dimensionality-reduction models, and GRU networks consistently outperform all benchmarks, with RMSFE reductions of roughly 40-60%; aggregation delivers further gains. Feature-attribution methods highlight industrial production, external trade, and labor-market indicators as dominant drivers of Singapore's short-run growth dynamics.

تم تطوير إطار جديد للتنبؤ في الوقت الحقيقي لتقييم نمو الناتج المحلي الإجمالي الفصلي في سنغافورة، باستخدام مجموعة عالية الأبعاد من حوالي 70 مؤشراً اقتصادياً ومالياً من الربع الأول من عام 1990 حتى الربع الثاني من عام 2023. يتضمن الإطار تقنيات متقدمة مثل الانحدارات المعاقبة، وخوارزميات التعلم الجماعي، والهياكل العصبية، مع ضمان ترتيب زمني واستخدام تحسين بايزي للمعلمات.

Se ha desarrollado un nuevo marco de previsión en tiempo real para evaluar el crecimiento del PIB trimestral de Singapur, utilizando un panel de alta dimensión de aproximadamente 70 indicadores económicos y financieros desde el primer trimestre de 1990 hasta el segundo trimestre de 2023. El marco incorpora técnicas avanzadas como regresiones penalizadas, algoritmos de aprendizaje en conjunto y arquitecturas neuronales, asegurando el orden temporal y empleando optimización bayesiana de hiperparámetros.

Un nouveau cadre de prévision en temps réel a été développé pour évaluer la croissance du PIB trimestriel de Singapour, utilisant un panel de haute dimension d'environ 70 indicateurs économiques et financiers de 1990T1 à 2023T2. Le cadre intègre des techniques avancées telles que les régressions pénalisées, les algorithmes d'apprentissage par ensemble et les architectures neuronales, tout en garantissant l'ordre temporel et en utilisant l'optimisation bayésienne des hyperparamètres.

A new real-time nowcasting framework has been developed to assess Singapore's quarterly GDP growth, utilizing a high-dimensional panel of approximately 70 economic and financial indicators from 1990Q1 to 2023Q2. The framework incorporates advanced techniques such as penalized regressions, ensemble learning algorithms, and neural architectures, while ensuring temporal ordering and employing Bayesian hyperparameter optimization.

Opening the Black Box: Nowcasting Singapore's GDP Growth and its Explainability

Was this article worth reading? Share it

Portfolio Backtest

Gonogo by Trampoline

Japan AI List