|
on Forecasting |
| By: | Boge Lyu; Qianye Yin; Iris Denise Tommelein; Hanyang Liu; Karnamohit Ranka; Karthik Yeluripati; Junzhe Shi |
| Abstract: | The persistent volatility of construction material prices poses significant risks to cost estimation, budgeting, and project delivery, underscoring the urgent need for granular and scalable forecasting methods. This study develops a forecasting framework that leverages the Construction Specifications Institute (CSI) MasterFormat as the target data structure, enabling predictions at the six-digit section level and supporting detailed cost projections across a wide spectrum of building materials. To enhance predictive accuracy, the framework integrates explanatory variables such as raw material prices, commodity indexes, and macroeconomic indicators. Four time-series models, Long Short-Term Memory (LSTM), Autoregressive Integrated Moving Average (ARIMA), Vector Error Correction Model (VECM), and Chronos-Bolt, were evaluated under both baseline configurations (using CSI data only) and extended versions with explanatory variables. Results demonstrate that incorporating explanatory variables significantly improves predictive performance across all models. Among the tested approaches, the LSTM model consistently achieved the highest accuracy, with RMSE values as low as 1.390 and MAPE values of 0.957, representing improvements of up to 59\% over the traditional statistical time-series model, ARIMA. Validation across multiple CSI divisions confirmed the framework's scalability, while Division 06 (Wood, Plastics, and Composites) is presented in detail as a demonstration case. This research offers a robust methodology that enables owners and contractors to improve budgeting practices and achieve more reliable cost estimation at the Definitive level. |
| Date: | 2025–12 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2512.09360 |
| By: | Eden Gross; Ryan Kruger; Francois Toerien |
| Abstract: | This study introduces a dynamic Bayesian network (DBN) framework for forecasting value at risk (VaR) and stressed VaR (SVaR) and compares its performance to several commonly applied models. Using daily S&P 500 index returns from 1991 to 2020, we produce 10-day 99% VaR and SVaR forecasts using a rolling period and historical returns for the traditional models, while three DBNs use both historical and forecasted returns. We evaluate the models' forecasting accuracy using standard backtests and forecasting error measures. Results show that autoregressive models deliver the most accurate VaR forecasts, while the DBNs achieve comparable performance to the historical simulation model, despite incorporating forward-looking return forecasts. For SVaR, all models produce highly conservative forecasts, with minimal breaches and limited differentiation in accuracy. While DBNs do not outperform traditional models, they demonstrate feasibility as a forward-looking approach to provide a foundation for future research on integrating causal inference into financial risk forecasting. |
| Date: | 2025–12 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2512.05661 |
| By: | A. L. Paredes |
| Abstract: | We examine the predictive power of a novel hybrid A3T-GCN architecture for forecasting closing stock prices of FTSE100 constituents. The dataset comprises 79 companies and 375, 329 daily observations from 2007 to 2024, with node features including technical indicators (RSI, MACD), normalized and log returns, and annualized log returns over multiple windows (ALR1W, ALR2W, ALR1M, ALR2M). Graphs are constructed based on sector classifications and correlations of returns or financial ratios. Our results show that the A3T-GCN model using annualized log-returns and shorter sequence lengths improves prediction accuracy while reducing computational requirements. Additionally, longer historical sequences yield only modest improvements, highlighting their importance for longer-term forecasts. |
| Date: | 2025–11 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2511.21873 |
| By: | Liang, Weifang; Liu, Yong; Somogyi, Simon; Anderson, David P. |
| Keywords: | Research Methods/Statistical Methods |
| Date: | 2024 |
| URL: | https://d.repec.org/n?u=RePEc:ags:aaea24:343687 |
| By: | Xiang Gao; Cody Hyndman |
| Abstract: | We develop an arbitrage-free deep learning framework for yield curve and bond price forecasting based on the Heath-Jarrow-Morton (HJM) term-structure model and a dynamic Nelson-Siegel parameterization of forward rates. Our approach embeds a no-arbitrage drift restriction into a neural state-space architecture by combining Kalman, extended Kalman, and particle filters with recurrent neural networks (LSTM/CLSTM), and introduces an explicit arbitrage error regularization (AER) term during training. The model is applied to U.S. Treasury and corporate bond data, and its performance is evaluated for both yield-space and price-space predictions at 1-day and 5-day horizons. Empirically, arbitrage regularization leads to its strongest improvements at short maturities, particularly in 5-day-ahead forecasts, increasing market-consistency as measured by bid-ask hit rates and reducing dollar-denominated prediction errors. |
| Date: | 2025–11 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2511.17892 |
| By: | Eghbal Rahimikia; Hao Ni; Weiguan Wang |
| Abstract: | Financial time series forecasting is central to trading, portfolio optimization, and risk management, yet it remains challenging due to noisy, non-stationary, and heterogeneous data. Recent advances in time series foundation models (TSFMs), inspired by large language models, offer a new paradigm for learning generalizable temporal representations from large and diverse datasets. This paper presents the first comprehensive empirical study of TSFMs in global financial markets. Using a large-scale dataset of daily excess returns across diverse markets, we evaluate zero-shot inference, fine-tuning, and pre-training from scratch against strong benchmark models. We find that off-the-shelf pre-trained TSFMs perform poorly in zero-shot and fine-tuning settings, whereas models pre-trained from scratch on financial data achieve substantial forecasting and economic improvements, underscoring the value of domain-specific adaptation. Increasing the dataset size, incorporating synthetic data augmentation, and applying hyperparameter tuning further enhance performance. |
| Date: | 2025–11 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2511.18578 |
| By: | Shilong Han |
| Abstract: | This study examines how China's high-speed rail (HSR) expansion affects analyst earnings forecast errors from an economic information friction perspective. Using firm-year panel data from 2008-2019, a period that covers HSR's early introduction and rapid nationwide rollout, the findings show that analysts' relative earnings forecast errors (RFE) decline significantly only after firms' cities become connected by high-speed rail. The placebo test, which artificially shifts HSR connectivity 3 years earlier than the actual opening year, yields an insignificant DID coefficient, rejecting the possibility that forecast errors were improving before the infrastructure shock. This supports the conclusion that forecast error reduction is linked to real geographic accessibility improvements rather than coincidence, pre-existing trends, or analyst anticipation. Economically, the study highlights that HSR reduces analysts' costs of gathering private, incremental information, particularly soft information obtained via plant or management visits. The rail network does not directly alter firms' internal capital allocation or earnings generation paths, but it lowers spatial barriers to information collection, enabling analysts to update EPS expectations under reduced travel friction. This work provides intuitive evidence that geography and mobility improvements contribute to forecasting accuracy in China's emerging, decentralized capital market corridors, and it encourages future research to consider transport accessibility as an exogenous information cost shock rather than an internal firm-capital shock. |
| Date: | 2025–12 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2512.03709 |
| By: | Luca Attolico |
| 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. |
| Date: | 2025–12 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2512.02092 |
| By: | Frederik Rech (School of Economics, Beijing Institute of Technology, Beijing, China); Fanchen Meng (Faculty of Economics, Shenzhen MSU-BIT University, Shenzhen, China); Hussam Musa (Faculty of Economics, Matej Bel University, Bansk\'a Bystrica, Slovakia); Martin \v{S}ebe\v{n}a (Faculty of Arts and Social Sciences, Hong Kong Baptist University, Hong Kong, China); Siele Jean Tuo (Business School, Liaoning University, Shenyang, China) |
| Abstract: | This study investigates whether firm-level artificial intelligence (AI) adoption improves the out-of-sample prediction of corporate financial distress models beyond traditional financial ratios. Using a sample of Chinese listed firms (2008-2023), we address sparse AI data with a novel pruned training window method, testing multiple machine learning models. We find that AI adoption consistently increases predictive accuracy, with the largest gains in recall rates for identifying distressed firms. Tree-based models and AI density metrics proved most effective. Crucially, models using longer histories outperformed those relying solely on recent "AI-rich" data. The analysis also identifies divergent adoption patterns, with healthy firms exhibiting earlier and higher AI uptake than distressed peers. These findings, while based on Chinese data, provide a framework for early-warning signals and demonstrate the broader potential of AI metrics as a stable, complementary risk indicator distinct from traditional accounting measures. |
| Date: | 2025–12 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2512.02510 |
| By: | Brian Ezinwoke; Oliver Rhodes |
| Abstract: | Modern high-frequency trading (HFT) environments are characterized by sudden price spikes that present both risk and opportunity, but conventional financial models often fail to capture the required fine temporal structure. Spiking Neural Networks (SNNs) offer a biologically inspired framework well-suited to these challenges due to their natural ability to process discrete events and preserve millisecond-scale timing. This work investigates the application of SNNs to high-frequency price-spike forecasting, enhancing performance via robust hyperparameter tuning with Bayesian Optimization (BO). This work converts high-frequency stock data into spike trains and evaluates three architectures: an established unsupervised STDP-trained SNN, a novel SNN with explicit inhibitory competition, and a supervised backpropagation network. BO was driven by a novel objective, Penalized Spike Accuracy (PSA), designed to ensure a network's predicted price spike rate aligns with the empirical rate of price events. Simulated trading demonstrated that models optimized with PSA consistently outperformed their Spike Accuracy (SA)-tuned counterparts and baselines. Specifically, the extended SNN model with PSA achieved the highest cumulative return (76.8%) in simple backtesting, significantly surpassing the supervised alternative (42.54% return). These results validate the potential of spiking networks, when robustly tuned with task-specific objectives, for effective price spike forecasting in HFT. |
| Date: | 2025–12 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2512.05868 |
| By: | Luca Attolico |
| Abstract: | Macroeconomic nowcasting sits at the intersection of traditional econometrics, data-rich information systems, and AI applications in business, economics, and policy. Machine learning (ML) methods are increasingly used to nowcast quarterly GDP growth, but adoption in high-stakes settings requires that predictive accuracy be matched by interpretability and robust uncertainty quantification. This article reviews recent developments in macroeconomic nowcasting and compares econometric benchmarks with ML approaches in data-rich and shock-prone environments, emphasizing the use of nowcasts as decision inputs rather than as mere error-minimization exercises. The discussion is organized along three axes. First, we contrast penalized regressions, dimension-reduction techniques, tree ensembles, and neural networks with autoregressive models, Dynamic Factor Models, and Random Walks, emphasizing how each family handles small samples, collinearity, mixed frequencies, and regime shifts. Second, we examine explainability tools (intrinsic measures and model-agnostic XAI methods), focusing on temporal stability, sign coherence, and their ability to sustain credible economic narratives and nowcast revisions. Third, we analyze non-parametric uncertainty quantification via block bootstrapping for predictive intervals and confidence bands on feature importance under serial dependence and ragged edge. We translate these elements into a reference workflow for "decision-grade" nowcasting systems, including vintage management, time-aware validation, and automated reliability audits, and we outline a research agenda on regime-dependent model comparison, bootstrap design for latent components, and temporal stability of explanations. Explainable ML and uncertainty quantification emerge as structural components of a responsible forecasting pipeline, not optional refinements. |
| Date: | 2025–11 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2512.00399 |
| By: | Jun Kevin; Pujianto Yugopuspito |
| Abstract: | This paper introduces a hybrid framework for portfolio optimization that fuses Long Short-Term Memory (LSTM) forecasting with a Proximal Policy Optimization (PPO) reinforcement learning strategy. The proposed system leverages the predictive power of deep recurrent networks to capture temporal dependencies, while the PPO agent adaptively refines portfolio allocations in continuous action spaces, allowing the system to anticipate trends while adjusting dynamically to market shifts. Using multi-asset datasets covering U.S. and Indonesian equities, U.S. Treasuries, and major cryptocurrencies from January 2018 to December 2024, the model is evaluated against several baselines, including equal-weight, index-style, and single-model variants (LSTM-only and PPO-only). The framework's performance is benchmarked against equal-weighted, index-based, and single-model approaches (LSTM-only and PPO-only) using annualized return, volatility, Sharpe ratio, and maximum drawdown metrics, each adjusted for transaction costs. The results indicate that the hybrid architecture delivers higher returns and stronger resilience under non-stationary market regimes, suggesting its promise as a robust, AI-driven framework for dynamic portfolio optimization. |
| Date: | 2025–11 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2511.17963 |
| By: | Drin, Svitlana (Örebro University School of Business); Avdieienko, Ivan (National University of Kyiv-Mohyla Academy); Chornei, Ruslan (National University of Kyiv-Mohyla Academy) |
| Abstract: | Modern companies face immense pressure to accelerate and refine decisions related to product assortment due to rapid changes and growing competition in the retail landscape. The volume, velocity, and volatility of business data make intuitive or situational approaches insufficient. Advances in optimization theory and forecasting models enable the design of robust, flexible decision-support systems that bridge the gap between business intuition and data-driven strategy. In retail, risk manifests primarily through operational inefficiencies, such as capital immobilized in unsold inventory and delayed responsiveness to demand changes. This demands a rethinking of risk modeling tailored specifically to the retail domain. At the same time, simplistic forecasting tools often prioritize short-term fluctuations at the expense of strategic seasonal trends, thereby undermining long-term planning. As a result, there is a critical need for integrated models that combine predictive accuracy with optimization under uncertainty. Such models must not only capture patterns in consumer demand but also align with operational constraints to ensure that solutions are implementable in practice. This work proposes a novel, multi-layered framework for assortment optimization that incorporates two key components: SARIMAX-based demand forecasting and the Discrete Functional Particle Method (DFPM) for iterative optimization. Additionally, we introduce a new operational risk measure Inventory Efficiency Ratio (IER) designed to quantify inefficiencies in the retail pipeline. By embedding these techniques into a unified system, we offer a practical solution for improving capital productivity, reducing inventory holding costs, and enhancing responsiveness in assortment planning. The methodology is validated through realworld data and demonstrates substantial performance improvements over standard planning strategies. |
| Keywords: | Retail assortment planning; SARIMAX forecasting; DFPM; inventory efficiency ratio; operational risk optimisation |
| JEL: | C61 |
| Date: | 2025–12–08 |
| URL: | https://d.repec.org/n?u=RePEc:hhs:oruesi:2025_017 |
| By: | Angelini, Elena; Darracq Pariès, Matthieu; Haertel, Thomas; Lalik, Magdalena; Aldama, Pierre; Brázdik, František; Damjanović, Milan; Fantino, Davide; Sanchez, Pablo Garcia; Guarda, Paolo; Kearney, Ide; Mociunaite, Laura; Saliba, Maria Christine; Sun, Yiqiao; Tóth, Máté Barnabás; Stoevsky, Grigor; Van der Veken, Wouter; Virbickas, Ernestas; Bulligan, Guido; Castro, Gabriela; Feješ, Martin; Grejcz, Kacper; Hertel, Katarzyna; Imbrasas, Darius; Kontny, Markus; Krebs, Bob; Opmane, Ieva; Rapa, Abigail Marie; Sariola, Mikko; Sequeira, Ana; Duarte, Rubén Veiga; Viertola, Hannu; Vondra, Klaus |
| Abstract: | This report provides a comprehensive overview of the models and tools used for macroeconomic projections within the European System of Central Banks (ESCB). These include semi-structural models, dynamic stochastic general equilibrium (DSGE) models, time series models and specialised satellite models tailored to particular questions or country-specific aspects. Each type of model has its own strengths and weaknesses and can help answer different questions. The models should therefore be seen as complementary rather than mutually exclusive. Semi-structural models are commonly used to produce baseline projection exercises, since they offer the flexibility to combine expert judgement with empirical data and have enough complexity and structure to provide a good representation of the economy. DSGE models, valued for their internal consistency and strong theoretical foundations, are another core forecasting tool used by some central banks, particularly to analyse counterfactuals. Time series models tend to be better suited to forecasting the short term, while scenario analysis and special events may require satellite models, extensions of existing models or even the development of new models tailored to the question at hand. The report also addresses the challenges to macroeconomic projections posed by data quality, including revisions and missing data, and describes the methods implemented to mitigate their effects. The report identifies “quick wins” to improve the projection process by enhancing the transparency and comparability of results through standardised reporting frameworks and better measurement of the judgement integrated in forecasts. The findings highlight the fundamental role of macroeconomic models in underpinning the ESCB’s projection exercises and ensuring that the Governing Council’s assessments and deliberations rest on coherent, granular and credible analysis of both demand-side and supply-side dynamics. JEL Classification: C30, C53, C54, E52 |
| Keywords: | economic models, forecasting, macroeconometrics, monetary policy |
| Date: | 2025–12 |
| URL: | https://d.repec.org/n?u=RePEc:ecb:ecbops:2025381 |
| By: | Bagattini Alexander; Chen Shao |
| Abstract: | Time series prediction algorithms are increasingly central to decision-making in high-stakes domains such as healthcare, energy management, and economic planning. Yet, these systems often inherit and amplify biases embedded in historical data, flawed problem specifications, and socio-technical design decisions. This paper critically examines the ethical foundations and mitigation strategies for algorithmic bias in time series prediction. We outline how predictive models, particularly in temporally dynamic domains, can reproduce structural inequalities and emergent discrimination through proxy variables and feedback loops. The paper advances a threefold contribution: First, it reframes algorithmic bias as a socio- technical phenomenon rooted in normative choices and institutional constraints. Second, it offers a structured diagnosis of bias sources across the pipeline, emphasizing the need for causal modeling, interpretable systems, and inclusive design practices. Third, it advocates for structural reforms that embed fairness through participatory governance, stakeholder engagement, and legally enforceable safeguards. Special attention is given to fairness validation in dynamic environments, proposing multi-metric, temporally-aware, and context- sensitive evaluation methods. Ultimately, we call for an integrated ethics-by-design approach that positions fairness not as a trade-off against performance, but as a co-requisite of responsible innovation. This framework is essential to developing predictive systems that are not only effective and adaptive but also aligned with democratic values and social equity. |
| Date: | 2025–12 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2512.01877 |
| By: | Klakow Akepanidtaworn; Korkrid Akepanidtaworn |
| Abstract: | Are Machine Learning (ML) algorithms superior to traditional econometric models for GDP nowcasting in a time series setting? Based on our evaluation of all models from both classes ever used in nowcasting across simulation and six country cases, traditional econometric models tend to outperform ML algorithms. Among the ML algorithms, linear ML algorithm – Lasso and Elastic Net – perform best in nowcasting, even surpassing traditional econometric models in cases of long GDP data and rich high-frequency indicators. Among the traditional econometric models, the Bridge and Dynamic Factor deliver the strongest empirical results, while Three-Pass Regression Filter performs well in our simulation. Due to the relatively short length of GDP series, complex and non-linear ML algorithms are prone to overfitting, which compromises their out-of-sample performance. |
| Keywords: | Nowcasting; Machine Learning; Forecast evaluation; Real-time data |
| Date: | 2025–12–05 |
| URL: | https://d.repec.org/n?u=RePEc:imf:imfwpa:2025/252 |
| By: | McWilliams, William N.; Isengildina Massa, Olga; Stewart, Shamar L. |
| Keywords: | Demand and Price Analysis, Agricultural and Food Policy, Risk and Uncertainty |
| Date: | 2024 |
| URL: | https://d.repec.org/n?u=RePEc:ags:aaea24:343923 |
| By: | Andersson, Jonas (Dept. of Business and Management Science, Norwegian School of Economics); Karlis, Dimitris (Dept. of Statistics, Athens University of Economics and Business) |
| Abstract: | The literature on multivariate time series is, largely, limited to either models based on the multivariate Gaussian distribution or models specifically developed for a given application. In this paper we develop a general approach which is based on an underlying, unobserved, Gaussian Vector Autoregressive (VAR) model. Using a transformation, we can capture the time dynamics as well as the distributional properties of a multivariate time series. The model is called the Vector AutoRegressive To Anyting (VARTA) model and was originally presented by Biller and Nelson (2003) who used it for the purpose of simulation. In this paper we derive a maximum likelihood estimator for the model and investigate its performance. We also provide diagnostic analysis and how to compute the predictive distribution. The proposed approach can provide better estimates about the forecasting distributions which can be of every kind not necessarily Gaussian distributions as for the standard VAR models. |
| Keywords: | non-Gaussian time series; maximum likelihood estimation; predictive distribution |
| JEL: | C13 C22 C58 |
| Date: | 2025–12–04 |
| URL: | https://d.repec.org/n?u=RePEc:hhs:nhhfms:2025_025 |
| By: | Tarek Jouini (Department of Economics, University of Windsor) |
| Abstract: | We propose an upper bound for the asymptotic approximation of the one-step-ahead forecast mean squared error (MSE) in infinite-order vector autoregression (VAR) settings, i.e., VAR(infinity). Once minimized over a truncation-lag of small order o(T^(1/3)), where T is the sample size, it yields a consistent truncation of the autoregression associated with the efficient one-step forecast error covariance matrix. When the infinite-order process degenerates to a finite-order VAR, we show that the resulting truncation is strongly consistent (eventually asymptotically), given a parameter epsilon >= 2. We particularly note that when epsilon tends to infinity, our order-selection criterion (upper bound) becomes inconsistent, with a variant of it reducing to Akaike information criterion (AIC). Thus, unlike the final prediction error (FPE) criterion and AIC, our criteria have the good sampling property of being consistent, like those by Hannan and Quinn, and Schwarz, respectively. Compared to conventional criteria, our model-selection procedures not only better handle the multivariate dynamic structure of the time series data, through a compound penalty term that we specify, but also tend to avoid model overfitting in large samples, hence the singularity problems encountered in practice. Variants of our primary criterion, which are in small samples less parsimonious than AIC in large systems, are also proposed. Besides being strongly consistent asymptotically, they tend to select the actual data-generating process (DGP) most of the time in small samples, as shown with Monte Carlo (MC) simulations. |
| Keywords: | infinite-order autoregression, truncation-lag, order-selection criterion, time series, strongly consistent asymptotically, Monte Carlo simulation. |
| JEL: | C13 C14 C15 C18 C22 C24 C32 C34 C51 C52 C53 C62 C63 C82 C83 |
| Date: | 2025–12 |
| URL: | https://d.repec.org/n?u=RePEc:wis:wpaper:2506 |