|
on Forecasting |
| By: | Katarzyna Chec; Bartosz Uniejewski; Rafal Weron |
| Abstract: | Electricity demand forecasts are crucial for power system operations. Market participants frequently rely on day-ahead predictions provided by Transmission System Operators (TSOs), but these can be systematically biased and - as recent studies report - may be improved using parsimonious autoregressive models. Despite the fact that many operational and economic decisions require well-calibrated uncertainty estimates, previous work has focused on point forecasts. The key question is how to derive accurate quantile and density predictions. Here we show that processing TSO forecasts with the Least Absolute Shrinkage and Selection Operator (LASSO) brings further accuracy gains and provides strong inputs for probabilistic forecasts. Drawing on ten years of data (2016-2025) from three European and North American power markets, we find that Generalized Additive Models for Location, Scale, and Shape (GAMLSS) deliver consistently better probabilistic performance than commonly used econometric and machine learning approaches. Together, these findings highlight how regularization and flexible distributional modeling can improve uncertainty quantification of electricity demand. |
| Keywords: | Electricity demand; Day-ahead market; LASSO; Probabilistic forecasting; GAMLSS |
| JEL: | C22 C45 C51 C52 C53 Q41 Q47 |
| Date: | 2026 |
| URL: | https://d.repec.org/n?u=RePEc:ahh:wpaper:worms2601 |
| By: | Mohammad Al Ridhawi; Mahtab Haj Ali; Hussein Al Osman |
| Abstract: | Stock market prediction presents considerable challenges for investors, financial institutions, and policymakers operating in complex market environments characterized by noise, non-stationarity, and behavioral dynamics. Traditional forecasting methods often fail to capture the intricate patterns and cross-sectional dependencies inherent in financial markets. This paper presents an integrated framework combining a node transformer architecture with BERT-based sentiment analysis for stock price forecasting. The proposed model represents the stock market as a graph structure where individual stocks form nodes and edges capture relationships including sectoral affiliations, correlated price movements, and supply chain connections. A fine-tuned BERT model extracts sentiment from social media posts and combines it with quantitative market features through attention-based fusion. The node transformer processes historical market data while capturing both temporal evolution and cross-sectional dependencies among stocks. Experiments on 20 S&P 500 stocks spanning January 1982 to March 2025 demonstrate that the integrated model achieves a mean absolute percentage error (MAPE) of 0.80% for one-day-ahead predictions, compared to 1.20% for ARIMA and 1.00% for LSTM. Sentiment analysis reduces prediction error by 10% overall and 25% during earnings announcements, while graph-based modeling contributes an additional 15% improvement by capturing inter-stock dependencies. Directional accuracy reaches 65% for one-day forecasts. Statistical validation through paired t-tests confirms these improvements (p |
| Date: | 2026–03 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2603.05917 |
| By: | Abhinav Das (Universität Ulm (Germany, Ulm)); Stephan Schlüter (Technische Hochschule Ulm (Germany, Ulm)); Lorenz Schneider (EM - EMLyon Business School) |
| Abstract: | This paper presents a new hybrid model for predicting German electricity prices. The algorithm is based on a combination of Gaussian process regression (GPR) and support vector regression (SVR). Although GPR is a competent model for learning sto-chastic patterns within data and for interpolation, its performance for out-of-sample data is not very promising. By choosing a suitable data-dependent covariance function, we enhance the performance of GPR. However, since the out-of-sample prediction is dependent on the training data, the prediction is vulnerable to noise and outliers. To overcome this issue, a separate prediction is calculated using SVR, which applies margin-based optimization. This method is advantageous when dealing with nonlinear processes and outliers, since only certain necessary points (support vectors) in the training data are responsible for regression. The individual predictions are then linearly combined using uniform weights. We evaluate the method on historical German day-ahead prices (2021–2023) and show that it outperforms publicly available benchmarks, namely, the LASSO estimated autore-gressive regression model and the deep neural network benchmark from the recent literature. |
| Keywords: | electricity price prediction, Gaussian process regression, support vector regression |
| Date: | 2026–02–24 |
| URL: | https://d.repec.org/n?u=RePEc:hal:journl:hal-05531916 |
| By: | Batuhan Koyuncu; Byeungchun Kwon; Marco Jacopo Lombardi; Fernando Perez-Cruz; Hyun Song Shin |
| Abstract: | This article introduces the BIS Time-series Regression Oracle (BISTRO), a general purpose time series model for macroeconomic forecasting. Its edge over traditional econometric approaches lies in its ability to deal with generic unconditional and conditional forecasting tasks, without requiring to adjust the model to the macroe conomic tasks being tackled. Building on the transformer architecture underlying LLMs, BISTRO is fine-tuned on the large repository of macroeconomic data main tained at the BIS. We show that BISTRO provides reliable unconditional forecasts for key macroeconomic aggregates and illustrate how using it for conditional fore casting can help unveiling patterns of nonlinearity in the data. |
| Keywords: | forecasting, scenarios, large language models |
| JEL: | C32 C45 C55 C87 |
| Date: | 2026–03 |
| URL: | https://d.repec.org/n?u=RePEc:bis:biswps:1337 |
| By: | Grewenig, Elisabeth (KfW, Frankfurt); Gründler, Klaus (University of Kassel); Lergetporer, Philipp (Technical University of Munich); Potrafke, Niklas (ifo Institute); Werner, Katharina (Business School Pforzheim); Zeidler, Helen (Technical University of Munich) |
| Abstract: | Public support for policy interventions depends on citizens’ beliefs about their likely effects. We examine how individuals form such beliefs by studying their predictions of experimental outcomes in a policy-relevant setting, and why their predictions differ from expert benchmarks. We elicit forecasts from 127 professional economists and a representative sample of 6, 200 German households about a large-scale behavioral experiment on education policy (N = 3, 133). Nonexperts predict both average outcomes and treatment effects far less accurately than experts. Prediction accuracy improves with calibrated priors, self-reported effort, and the use of structured reasoning, but remains well below expert levels. We show that scalable design features, including the provision of well-calibrated numerical anchors and monetary incentives to rise effort, improve non-expert predictions, with effects comparable in magnitude to tertiary education or structured reasoning. Our findings have important implications for bridging the ‘expertise gap’ in public discourse. |
| Keywords: | expert forecasts, lay predictions, belief formation, expertise gap, policy support, behavioral experiments |
| JEL: | A11 D83 H52 I22 |
| Date: | 2026–03 |
| URL: | https://d.repec.org/n?u=RePEc:iza:izadps:dp18453 |
| By: | Colin Aitken; Rajat Masiwal; Adam Marchakitus; Katherine Kowal; Mayank Gupta; Tyler Yang; Amir Jina; Pedram Hassanzadeh; William R. Boos; Michael Kremer |
| Abstract: | Hundreds of millions of farmers make high-stakes decisions under uncertainty about future weather. Forecasts can inform these decisions, but available choices and their risks and benefits vary between farmers. We introduce a decision-theory framework for designing useful forecasts in settings where the forecaster cannot prescribe optimal actions because farmers' circumstances are heterogeneous. We apply this framework to the case of seasonal onset of monsoon rains, a key date for planting decisions and agricultural investments in many tropical countries. We develop a system for tailoring forecasts to the requirements of this framework by blending systematically benchmarked artificial intelligence (AI) weather prediction models with a new "evolving farmer expectations" statistical model. This statistical model applies Bayesian inference to historical observations to predict time-varying probabilities of first-occurrence events throughout a season. The blended system yields more skillful Indian monsoon forecasts at longer lead times than its components or any multi-model average. In 2025, this system was deployed operationally in a government-led program that delivered subseasonal monsoon onset forecasts to 38 million Indian farmers, skillfully predicting that year's early-summer anomalous dry period. This decision-theory framework and blending system offer a pathway for developing climate adaptation tools for large vulnerable populations around the world. |
| Date: | 2026–03 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2603.07893 |
| By: | Harrison Katz |
| Abstract: | Tourism demand forecasting is methodologically mature, but it typically treats accommodation supply as fixed or exogenous. In platform-mediated short-term rentals, supply is elastic, decision-driven, and co-evolves with demand through pricing, information design, and interventions. I reframe the core issue as endogenous stock-out censoring: realized booked nights satisfy B_{k, t} |
| Date: | 2026–02 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2603.00422 |
| By: | Ceballos, Francisco; Hernandez, Manuel A. |
| Abstract: | The external and internal Migration Propensity Indices (e-MPI and i-MPI) are tools to objectively estimate the probability that individuals from a given household will, respectively, migrate abroad or migrate domestically in the near future. We use new longitudinal data to test their predictive performance fully out of sample. We find that households classified as being of high-propensity to migrate by the e-MPI were significantly more likely to migrate abroad within 24 months (10.7%) than medium- and low-propensity groups (8.8% and 5.3%). For domestic migration, the i-MPI shows an even stronger gradient (19.6% versus 8.5% and 3.5%, respectively). Regression models confirm that both indices outperform alternative predictors—including income, climate shocks, crime, and migration intent—and maintain predictive power across rural and urban areas. Placebo tests indicate that the e-MPI and i-MPI capture distinct dimensions of migration behavior, validating their use for targeting and monitoring migration-related interventions. Overall, the MPIs emerge as simple yet statistically robust tools that reliably predict both international and domestic migration, offering a practical and scalable solution to help governments and development agencies anticipate migration trends and allocate resources strategically. |
| Keywords: | migration; indicators; forecasting; population dynamics; Honduras; Central America |
| Date: | 2025–12–17 |
| URL: | https://d.repec.org/n?u=RePEc:fpr:ifprwp:178954 |
| By: | Zongwu Cai (Department of Economics, The University of Kansas, Lawrence, KS 66045, USA); Yifeng Chen (Department of Economics, Nanyang Technological University, Singapore 639798); Seok Young Hong (Department of Economics, Nanyang Technological University, Singapore 639798); Daniel Tsvetanov (Norwich Business School, University of East Anglia, Norwich NR4 7TJ, UK) |
| Abstract: | We develop an empirical likelihood framework for testing return predictability in the conditional mean and conditional quantiles. A unified chi-square limit theory is established across a broad spectrum of predictor persistence, including stationary, mildly integrated, nearly integrated, unit-root, and mildly explosive cases. We provide two complementary approaches to handle the unknown intercept: (i) a sample-splitting approach under relaxed regularity conditions and (ii) a new two-stage method that improves efficiency and accommodates quantile inference, where sample-splitting is infeasible. We examine the finite-sample bias of the two-stage method, and propose a bias-correction scheme and gradually saturated weights that improve performance under high persistence. Simulation evidence demonstrates that our tests exhibit competitive size and power across persistence classes, with notable gains in quantile predictability. An empirical application to the U.S. stock market shows modest evidence of mean predictability, whereas quantile-based inference reveals stronger and economically relevant predictability in the tails of the return distribution. |
| Keywords: | Predictive Mean Regression; Predictive Quantile Regression; Empirical Likelihood; Bartlett Bias Correction. |
| JEL: | C12 C32 C51 C52 |
| Date: | 2026–01 |
| URL: | https://d.repec.org/n?u=RePEc:kan:wpaper:202609 |
| By: | Sudipta Basu; Xin Dai; Caroline Lee |
| Abstract: | This paper examines how multi-unit firms’ life-cycle stages affect analyst forecast accuracy. While prior studies focus on the firm-level life cycle, we utilize the Census data and focus on the establishment level. We find that analyst forecast accuracy is lower for multi-unit firms whose establishments are in different life-cycle stages than those in the same life-cycle stage. This finding suggests that the forecasting difficulty of more diversified firms can be attributed to the different life-cycle stages of each establishment. We also find that for firms whose units are in different stages, analyst forecast accuracy is lower if the establishments in earlier stages are larger (i.e., generate more revenue) than those in later stages. As a comparison, we estimate the life-cycle stages using firms’ segment classifications in their 10-K filings. We find that analysts’ forecast accuracy is lower when firms report fewer segments than the number of establishments, suggesting that aggregating more establishments for segment reporting could complicate analysts’ forecasting. To our knowledge, this is the first study that focuses on the establishment-level life cycle. This study highlights that firm-level life cycles should not be taken without caution, as aggregating multiple units’ life cycles may be misleading. In order to provide better forecasts to investors, analysts should have a deeper understanding of firms’ subunits, especially when the establishments are in different life-cycle stages. |
| Date: | 2026–02 |
| URL: | https://d.repec.org/n?u=RePEc:cen:wpaper:26-12 |
| By: | Chris Jackson; Daniel Rivera Greenwood |
| Abstract: | This paper presents a nowcasting model for global trade that allows for regional dynamics and spillovers. World trade growth is driven by common global factors but also regional trends. While existing trade nowcasting models have focused on the former, we allow for the latter using a dynamic factor model (DFM) with a multi-factor block structure. By directly modeling global trends, regional variation and spillovers, we improve on the performance of standard trade nowcasting models, particularly periods characterized by regional heterogeneity. A multi-factor regional framework may be particularly advantageous for tracking trade developments in the future given a period changing trade patterns and geo-economic fragmentation. The model also sheds light on trade spillovers and the drivers of news in global trade: Asia, in particular, has notable spillovers to the global and other regional trade cycles. |
| Keywords: | Global trade; Nowcasting; Dynamic Factor Model; Spillovers |
| Date: | 2026–03–13 |
| URL: | https://d.repec.org/n?u=RePEc:imf:imfwpa:2026/048 |
| By: | Dobrislav Dobrev; Pawel J. Szerszén |
| Abstract: | Replacing erroneous observations with missing values is known to mitigate outlier-induced distortions in state-space model inference. Yet, in economic data, outliers can be small and difficult to detect, while still occurring in temporal clusters and generating persistent distortions. We therefore put forward an unsupervised approach for exogenously randomized substitution of missing data (RMDX), designed as an ensemble-averaging enhancement that can be used to improve the robustness of any filter also to more elusive outliers. Our bias-variance decomposition theory for RMDX ensemble averaging establishes that, under mild regularity conditions on the influence of outliers, the missing data randomization rate acts as a regularization parameter, which can be set optimally to minimize mean squared error loss using standard cross-validation. We corroborate these theoretical results using Monte Carlo simulations, which show that RMDX ensemble averaging can substantially enhance the performance of commonly used robust filters, including ones that rely on supervised missing data substitution upon exceeding outlier detection thresholds. As anticipated, the gains are most pronounced in the presence of patches of moderately sized outliers that are difficult to mitigate. To further assess empirical relevance in economics, we also document that RMDX-enhanced filters perform favorably in widely used state-space models for extracting inflation trends, where clusters of measurement outliers in inflation data are known to pose an extra challenge. |
| Keywords: | State-space models; outlier-robust filtering and forecasting; missing data randomization; bagging and ensemble averaging; bias-variance tradeoff. |
| JEL: | C15 C22 C53 E37 |
| Date: | 2026–03 |
| URL: | https://d.repec.org/n?u=RePEc:gwc:wpaper:2026-004 |