|
on Forecasting |
| By: | Ciaran O'Connor; Mohamed Bahloul; Steven Prestwich; Andrea Visentin |
| Abstract: | Electricity price forecasting has become a critical tool for decision-making in energy markets, particularly as the increasing penetration of renewable energy introduces greater volatility and uncertainty. Historically, research in this field has been dominated by point forecasting methods, which provide single-value predictions but fail to quantify uncertainty. However, as power markets evolve due to renewable integration, smart grids, and regulatory changes, the need for probabilistic forecasting has become more pronounced, offering a more comprehensive approach to risk assessment and market participation. This paper presents a review of probabilistic forecasting methods, tracing their evolution from Bayesian and distribution based approaches, through quantile regression techniques, to recent developments in conformal prediction. Particular emphasis is placed on advancements in probabilistic forecasting, including validity-focused methods which address key limitations in uncertainty estimation. Additionally, this review extends beyond the Day-Ahead Market to include the Intra-Day and Balancing Markets, where forecasting challenges are intensified by higher temporal granularity and real-time operational constraints. We examine state of the art methodologies, key evaluation metrics, and ongoing challenges, such as forecast validity, model selection, and the absence of standardised benchmarks, providing researchers and practitioners with a comprehensive and timely resource for navigating the complexities of modern electricity markets. |
| Date: | 2025–10 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2511.05523 |
| By: | Xu Zhang; Zhengang Huang; Yunzhi Wu; Xun Lu; Erpeng Qi; Yunkai Chen; Zhongya Xue; Qitong Wang; Peng Wang; Wei Wang |
| Abstract: | Time series forecasting is important in finance domain. Financial time series (TS) patterns are influenced by both short-term public opinions and medium-/long-term policy and market trends. Hence, processing multi-period inputs becomes crucial for accurate financial time series forecasting (TSF). However, current TSF models either use only single-period input, or lack customized designs for addressing multi-period characteristics. In this paper, we propose a Multi-period Learning Framework (MLF) to enhance financial TSF performance. MLF considers both TSF's accuracy and efficiency requirements. Specifically, we design three new modules to better integrate the multi-period inputs for improving accuracy: (i) Inter-period Redundancy Filtering (IRF), that removes the information redundancy between periods for accurate self-attention modeling, (ii) Learnable Weighted-average Integration (LWI), that effectively integrates multi-period forecasts, (iii) Multi-period self-Adaptive Patching (MAP), that mitigates the bias towards certain periods by setting the same number of patches across all periods. Furthermore, we propose a Patch Squeeze module to reduce the number of patches in self-attention modeling for maximized efficiency. MLF incorporates multiple inputs with varying lengths (periods) to achieve better accuracy and reduces the costs of selecting input lengths during training. The codes and datasets are available at https://github.com/Meteor-Stars/MLF. |
| Date: | 2025–11 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2511.08622 |
| By: | Yilong Zeng; Boyan Tang; Xuanhao Ren; Sherry Zhefang Zhou; Jianghua Wu; Raymond Lee |
| Abstract: | This paper introduces the Fractal-Chaotic Oscillation Co-driven (FCOC) framework, a novel paradigm for financial volatility forecasting that systematically resolves the dual challenges of feature fidelity and model responsiveness. FCOC synergizes two core innovations: our novel Fractal Feature Corrector (FFC), engineered to extract high-fidelity fractal signals, and a bio-inspired Chaotic Oscillation Component (COC) that replaces static activations with a dynamic processing system. Empirically validated on the S\&P 500 and DJI, the FCOC framework demonstrates profound and generalizable impact. The framework fundamentally transforms the performance of previously underperforming architectures, such as the Transformer, while achieving substantial improvements in key risk-sensitive metrics for state-of-the-art models like Mamba. These results establish a powerful co-driven approach, where models are guided by superior theoretical features and powered by dynamic internal processors, setting a new benchmark for risk-aware forecasting. |
| Date: | 2025–11 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2511.10365 |
| By: | Krzysztof Drachal (Faculty of Economic Sciences, University of Warsaw); Joanna J?drzejewska (Faculty of Economic Sciences, University of Warsaw) |
| Abstract: | Bayesian dynamic mixture models offer a flexible framework for capturing evolving relationships between dependent and independent variables over time. They address both structural and variable uncertainty, incorporating real-time market information through dynamic updating. Unlike static approaches, they allow the underlying process to change, which is particularly relevant for the fluctuating nature of commodity markets. In scenarios with a large number of possible predictors, various regression models can be employed, each yielding its own probability distribution for the coefficients. Forecasts are then constructed by combining these distributions using time-varying weights. This paper utilizes Bayesian dynamic mixture models to allow both the regression parameters and their associated weights to change over time. Computational efficiency is maintained by preserving distributional forms and limiting numerical approximations to statistics distributions. The study uses monthly Global Price Index of All Commodities from the International Monetary Fund, spanning the period 2003?2024. Key explanatory variables include interest rates, exchange rates, and stock market indices. The forecasting performance of the proposed models is compared to other techniques such as Dynamic Model Averaging, LASSO, ridge regression, and ARIMA, etc. Evaluation is conducted using the Diebold-Mariano test, Giacomini-Rossi test, Model Confidence Set procedure, and Clark-West test. (This research was funded in whole by National Science Centre, Poland, grant number 2022/45/B/HS4/00510.) |
| Keywords: | Bayesian dynamic mixture models; Commodities prices; Mixture models; Model averaging; Time-series forecasting; Variable uncertainty |
| JEL: | C32 C53 Q02 |
| URL: | https://d.repec.org/n?u=RePEc:sek:iefpro:15316933 |
| By: | Bachmair, K.; Schmitz, N. |
| Abstract: | While financial markets are known to contain information about future economic developments, the channels through which asset prices enhance macroeconomic forecastability remain insufficiently understood. We develop a structured set of like-for-like experiments to isolate which data and model properties drive forecasting power. Using U.S. data on inflation, industrial production, unemployment and equity returns, we test eight hypotheses along two dimensions: the contribution of financial data given different estimation methods and model classes, and the role of model choice given different financial inputs. Data aspects include cross-sectional granularity, intra-period frequency, and real-time, revisionless availability; model aspects include sparsity, direct versus indirect specification, nonlinearity, and state dependence on volatile periods. We find that financial data can deliver consistent and economically meaningful gains, but only under suitable modeling choices: Random Forest most reliably extracts useful signals, whereas an unregularised VAR often fails to do so; by contrast, expanding the financial information set along granularity, frequency, or real-time dimensions yields little systematic benefit. Gains strengthen somewhat under elevated policy uncertainty, especially for inflation, but are otherwise fragile. The analysis clarifies how data and model choices interact and provides practical guidance for forecasters on when and how to use financial inputs. |
| Keywords: | Macroeconomic Forecasting, Stock Returns, Hypothesis Testing, Machine Learning, Regularisation, Vector Autoregressions, Ridge Regression, Lasso, Random Forests, Support Vector Regression, Elastic Net, Principal Component Analysis, Neural Networks |
| JEL: | C32 C45 C53 C58 E27 E37 E44 G17 |
| Date: | 2025–11–13 |
| URL: | https://d.repec.org/n?u=RePEc:cam:camdae:2574 |
| By: | Sotiris; Tsolacos; Tatiana Franus |
| Abstract: | In this paper, we evaluate the performance of various methodologies for forecasting real estate yields. Expected yield changes are a crucial input for valuations and investment strategies. We conduct a comparative study to assess the forecast accuracy of econometric and time series models relative to machine learning algorithms. Our target series include net initial and equivalent yields across key real estate sectors: office, industrial, and retail. The analysis is based on monthly UK data, though the framework can be applied to different contexts, including quarterly data. The econometric and time series models considered include ARMA, ARMAX, stepwise regression, and VAR family models, while the machine learning methods encompass Random Forest, XGBoost, Decision Tree, Gradient Boosting and Support Vector Machines. We utilise a comprehensive set of economic, financial, and survey data to predict yield movements and evaluate forecast performance over three-, six-, and twelve-month horizons. While conventional forecast metrics are calculated, our primary focus is on directional forecasting. The findings have significant practical implications. By capturing directional changes, our assessment aids price discovery in real estate markets. Given that private-market real estate data are reported with a lag - even for monthly data - early signals of price movements are valuable for investors and lenders. This study aims to identify the most successful methods to gauge forthcoming yield movements. |
| Keywords: | directional forecasting; econometric models; Machine Learning; property yields |
| JEL: | R3 |
| Date: | 2025–01–01 |
| URL: | https://d.repec.org/n?u=RePEc:arz:wpaper:eres2025_269 |
| By: | Matteo Bonato (Department of Economics and Econometrics, University of Johannesburg, Auckland Park, South Africa; IPAG Business School, 184 Boulevard Saint-Germain, 75006 Paris, France; B-CCaS, University of Edinburgh Business School); Oguzhan Cepni (Ostim Technical University, Ankara, Turkiye; University of Edinburgh Business School, Centre for Business, Climate Change, and Sustainability; Department of Economics, Copenhagen Business School, Denmark); Rangan Gupta (Department of Economics, University of Pretoria, Private Bag X20, Hatfield 0028, South Africa); Christian Pierdzioch (Department of Economics, Helmut Schmidt University, Holstenhofweg 85, P.O.B. 700822, 22008 Hamburg, Germany) |
| Abstract: | We study the out-of-sample forecasting value of and state-level and market-wideoverall commercial, industrial, and residential electricity sales for monthly state-level (1995--2025) realized stock market volatility (RV) of the United States (U.S.). We control for state-level and market-wide realized moments (leverage, skewness, kurtosis, and tail risks). We estimate our forecasting models using a boosting algorithm, and two alternative statistical learning algorithms (forward best predictor selection and random forests). We find evidence that realized moments have predictive power for subsequent RV at forecast horizons up to one year in some model configurations, while evidence of predictive power of the growth rate of electricity sales, whether measured at state-level or at the market-level, is mixed and mainly concentrated, on average across states, at the short forecast horizon. |
| Keywords: | Stock market, Realized volatility, Electricity sales, Statistical learning, Forecasting |
| JEL: | C22 C53 G10 G17 Q41 |
| Date: | 2025–11 |
| URL: | https://d.repec.org/n?u=RePEc:pre:wpaper:202540 |
| By: | Gabriel M. Arantes; Richard F. Pinto; Bruno L. Dalmazo; Eduardo N. Borges; Giancarlo Lucca; Viviane L. D. de Mattos; Fabian C. Cardoso; Rafael A. Berri |
| Abstract: | Binary options trading is often marketed as a field where predictive models can generate consistent profits. However, the inherent randomness and stochastic nature of binary options make price movements highly unpredictable, posing significant challenges for any forecasting approach. This study demonstrates that machine learning algorithms struggle to outperform a simple baseline in predicting binary options movements. Using a dataset of EUR/USD currency pairs from 2021 to 2023, we tested multiple models, including Random Forest, Logistic Regression, Gradient Boosting, and k-Nearest Neighbors (kNN), both before and after hyperparameter optimization. Furthermore, several neural network architectures, including Multi-Layer Perceptrons (MLP) and a Long Short-Term Memory (LSTM) network, were evaluated under different training conditions. Despite these exhaustive efforts, none of the models surpassed the ZeroR baseline accuracy, highlighting the inherent randomness of binary options. These findings reinforce the notion that binary options lack predictable patterns, making them unsuitable for machine learning-based forecasting. |
| Date: | 2025–11 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2511.15960 |
| By: | Kelvin J. L. Koa; Jan Chen; Yunshan Ma; Huanhuan Zheng; Tat-Seng Chua |
| Abstract: | While Large Language Models have been used to produce interpretable stock forecasts, they mainly focus on analyzing textual reports but not historical price data, also known as Technical Analysis. This task is challenging as it switches between domains: the stock price inputs and outputs lie in the time-series domain, while the reasoning step should be in natural language. In this work, we introduce Verbal Technical Analysis (VTA), a novel framework that combine verbal and latent reasoning to produce stock time-series forecasts that are both accurate and interpretable. To reason over time-series, we convert stock price data into textual annotations and optimize the reasoning trace using an inverse Mean Squared Error (MSE) reward objective. To produce time-series outputs from textual reasoning, we condition the outputs of a time-series backbone model on the reasoning-based attributes. Experiments on stock datasets across U.S., Chinese, and European markets show that VTA achieves state-of-the-art forecasting accuracy, while the reasoning traces also perform well on evaluation by industry experts. |
| Date: | 2025–11 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2511.08616 |
| By: | Ramaharo, Franck M. |
| Abstract: | In this paper, we investigate the predictive power of petroleum consumption for Malagasy real GDP using the Mixed Data Sampling (MIDAS) framework over the period 2007-2024. While GDP data are available at a quarterly frequency, petroleum consumption is observed monthly and disaggregated by sectoral use and product type. We use this high-frequency disaggregated data to identify which components deliver the strongest nowcasting performance. Our results show that, at the sectoral level, transportation, aviation and bunkers consistently deliver the most accurate GDP nowcasts over the sample period. The best-performing product-level specifications correspond precisely to the fuels predominantly used in these sectors, namely, gas oil, super-unleaded petrol, aviation gasoline, and jet fuel. The aggregate measure of total petroleum consumption also yields competitive forecasting accuracy across specifications. This supports its use as a broad high-frequency indicator of economic activity. Our findings suggest that forecasters of Madagascar’s GDP can significantly improve predictive accuracy by using appropriately disaggregated energy data, particularly from sectoral categories linked to mobility and trade. |
| Keywords: | nowcasting; petroleum consumption; real gross domestic product; MIDAS; Mixed-frequency Data Sampling; Madagascar |
| JEL: | C53 E17 O47 Q43 |
| Date: | 2025–10–27 |
| URL: | https://d.repec.org/n?u=RePEc:pra:mprapa:126629 |
| By: | Escribano, Álvaro; Rodríguez, Juan Andrés |
| Abstract: | The year 2024 marked a critical milestone in global warming, with global mean temperatures exceeding pre-industrial levels by 1.55 °C. Polar ice loss, largely driven by anthropogenic CO₂ emissions, exhibits highly nonlinear dynamics that challenge conventional linear modeling approaches. This paper investigates the nonlinear effects of CO₂ on Arctic and Antarctic sea-ice volumes using regime-switching econometric models defined for both accumulated concentration levels and annual changes in CO₂. Specifically, we estimate reduced-form Smooth Transition Autoregressive (STAR) and Threshold Autoregressive (TAR) specifications using data generated from General Circulation Models (GCMs). We consider the identification of CO₂ thresholds and the evaluation of sea-ice dynamics under alternative CO₂ emissions trajectories, such as the IPCC’s Shared Socioeconomic Pathways (SSPs). Our results reveal important threshold effects across hemispheres. For the Arctic, a single CO₂ threshold is identified at approximately 330 ppm (or 𝛥𝐶𝑂2 = 1.16 ppm), while for the Antarctic, two thresholds emerge at 285 ppm (or 𝛥𝐶𝑂2 = 0.13 ppm) and 321 ppm (or 𝛥𝐶𝑂2 = 0.56). Beyond these points, the decline in sea-ice volume accelerates sharply. Forecasts under business-asusual (BAU) scenarios suggest that the Arctic could become ice-free around 2060 [2045, 2078], while Antarctic sea-ice loss may extend well beyond 2100. Under an intermediate emissions path, such as SSP2-4.5, recovery of sea-ice volume remains feasible if global CO₂ growth begins to decline after 2035 by an average of 3.2 ppm every year, with projected reversion to historical levels around 2075 for both hemispheres. |
| Keywords: | Climate change; Climate econometrics; Sea-ice volume; Shared socioeconomic pathways (SSPs); Smooth transition autoregressive (STAR) models; Threshold autoregressive (TAR) models; CO₂ concentration thresholds |
| JEL: | C5 C32 C38 C51 C52 |
| Date: | 2025–11–18 |
| URL: | https://d.repec.org/n?u=RePEc:cte:werepe:48471 |
| By: | Jan Schmid; He Cheng |
| Abstract: | The increasing complexities of real estate market forecasting, in combination with the accelerated evolution of machine learning (ML) algorithms, necessitates the optimisation of algorithm selection to reduce computational demands and enhance model accuracy. While numerous studies have examined the performance of individual algorithms, a significant research gap remains concerning the impact of dataset characteristics on algorithmic performance within this specific domain. The present study aims to address this research gap by undertaking a systematic meta-learning analysis of 54 real estate forecasting studies conducted between 2001 and 2024. The study explores the relationship between dataset characteristics and algorithm performance, focusing on factors such as dataset size, dimensionality, and variable categories. Two models, a decision tree and a random forest model, were utilised to assess the impact of these characteristics on the accuracy of various algorithm categories, including artificial neural networks (ANNs), ensemble methods, and support vector machines (SVMs).The study's findings suggest that the random forest algorithm, when applied to dataset characteristics, serves as a reliable tool for predicting the best-performing algorithm for a given real estate market forecasting dataset. The model attained an average area under the curve (AUC) of 0.98 and an overall accuracy of 88%, underscoring the practical relevance of meta-learning approaches in econometrics and highlighting the potential for further enhancing algorithm selection methodologies in this research domain.This research contributes to the expanding field of automated machine meta-learning by providing a framework for more efficient and accurate real estate market forecasting. |
| Keywords: | algorithm selection; meta-learning; Random forest; real estate forecasting |
| JEL: | R3 |
| Date: | 2025–01–01 |
| URL: | https://d.repec.org/n?u=RePEc:arz:wpaper:eres2025_40 |
| By: | Rustam Ibragimov; Jihyun Kim; Anton Skrobotov |
| Abstract: | This paper develops robust inference methods for predictive regressions that address key challenges posed by endogenously persistent or heavy-tailed regressors, as well as persistent volatility in errors. Building on the Cauchy estimation framework, we propose two novel tests: one based on $t$-statistic group inference and the other employing a hybrid approach that combines Cauchy and OLS estimation. These methods effectively mitigate size distortions that commonly arise in standard inference procedures under endogeneity, near nonstationarity, heavy tails, and persistent volatility. The proposed tests are simple to implement and applicable to both continuous- and discrete-time models. Extensive simulation experiments demonstrate favorable finite-sample performance across a range of realistic settings. An empirical application examines the predictability of excess stock returns using the dividend-price and earnings-price ratios as predictors. The results suggest that the dividend-price ratio possesses predictive power, whereas the earnings-price ratio does not significantly forecast returns. |
| Date: | 2025–11 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2511.09249 |
| By: | Jan Schmid; He Cheng; Francisco Amaral; Zdrzalek Jonas |
| Abstract: | This study proposes the conception of a real estate market forecasting model for the research area of Frankfurt am Main, utilising sophisticated AI algorithms to predict house price indices. The model integrates two primary datasets: listing data from ImmoScout24 and transaction data from the local expert committee. These datasets were linked using a threshold optimisation approach to ensure accurate matching of listings and transactions at the object level. A comprehensive review of prior studies was conducted to select key predictors, supplemented by novel variables that measure differences between listing and transaction data, such as price differences and time on market. The Random Forest based algorithm selection process involved a meta-learning approach of 54 prior studies, adapted to the final dataset's structure. The XGBoost model was selected as the most suitable algorithm, achieving a Mean Absolute Percentage Error (MAPE) of 1.76% and a Root Mean Square Error (RMSE) of 2.43 on the testing dataset. The methodology also incorporated macroeconomic and socio-economic indicators, with data structured into spatial-temporal grids for quarterly forecasting. The model demonstrated high predictive accuracy, offering valuable insights for real estate market analysis and future decision-making. Subsequent research is planned to validate the model and apply it to additional urban regions, initially focusing on the seven largest cities in Germany. |
| Keywords: | Forecasting; House Price Indices; Real Estate Market; XGBoost |
| JEL: | R3 |
| Date: | 2025–01–01 |
| URL: | https://d.repec.org/n?u=RePEc:arz:wpaper:eres2025_84 |
| By: | Chen Jin; Ankush Agarwal |
| Abstract: | We introduce a conditional Denoising Diffusion Probabilistic Model (DDPM) for generating arbitrage-free implied volatility (IV) surfaces, offering a more stable and accurate alternative to existing GAN-based approaches. To capture the path-dependent nature of volatility dynamics, our model is conditioned on a rich set of market variables, including exponential weighted moving averages (EWMAs) of historical surfaces, returns and squared returns of underlying asset, and scalar risk indicators like VIX. Empirical results demonstrate our model significantly outperforms leading GAN-based models in capturing the stylized facts of IV dynamics. A key challenge is that historical data often contains small arbitrage opportunities in the earlier dataset for training, which conflicts with the goal of generating arbitrage-free surfaces. We address this by incorporating a standard arbitrage penalty into the loss function, but apply it using a novel, parameter-free weighting scheme based on the signal-to-noise ratio (SNR) that dynamically adjusts the penalty's strength across the diffusion process. We also show a formal analysis of this trade-off and provide a proof of convergence showing that the penalty introduces a small, controllable bias that steers the model toward the manifold of arbitrage-free surfaces while ensuring the generated distribution remains close to the real-world data. |
| Date: | 2025–11 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2511.07571 |
| By: | Magaletti, Nicola; Nortarnicola, Valeria; Di Molfetta, Mauro; Mariani, Stefano; Leogrande, Angelo |
| Abstract: | This paper illustrates an original Data-Driven Analysis on logistics matters, developed within the scope of a research & development project conducted by La Logistica S.r.l. and financed by the Apulia Region. The paper uses real operative data for developing and validating a complete analytical tool to forecast volume saturation on pallets. Compared to other studies on similar matters, which frequently make use of simulated and ideal sets of data, this research provides a unique opportunity to test, on an empirical ground, how very advanced Data Science approaches could be successfully applied on industrial processes.The paper develops a complete analytical investigation combining diagnose analytics, forecasting and model-interpretability to compare the performances of many different Machine Learning algorithms, such us K-Nearest Neighbors, Decision Trees, Random Forest, Boosting, Support Vector Machines, and Linear Models. The results obtained after an overarching comparison show how KNN is significantly more accurate and trustful in relation to any other method, outperforming any other on any key error measure and interpretability factors. The addition of contribution and importance investigation on variables extends to this paper a unique degree of originality, showing which SKUs’ physical variables are most significantly influencing volume saturation on pallets, and offering immediate management implications related to their operative meaning. The complete methodological pipeline is grounded on real operative data regarding La Logistica S.r.l., making this paper capable to show how operative Data-Driven solutions could play an innovative role into improving operative continuity within logistics matters. |
| Date: | 2025–11–18 |
| URL: | https://d.repec.org/n?u=RePEc:osf:socarx:5aw8b_v1 |
| By: | Fatih Eren Metin; Kerem Yavuz Arslanli |
| Abstract: | The construction sector in Turkey is highly influenced by price fluctuations, economic crises, and external shocks. In particular, the major earthquake in 2023 accelerated the increase in construction costs, revealing that price volatility in Turkey is significantly higher compared to other OECD countries. This study develops a forecasting model for construction costs using the Vector Error Correction Model (VECM). A total of 16 periods of data were collected from the first quarter of 2021 to the last quarter of 2024. The selected variables include total cost, labor cost, material cost, iron prices, and the Consumer Price Index (CPI). The VECM model was applied to analyze long-term economic relationships and short- term fluctuations in construction costs. The results were compared with traditional forecasting methods, demonstrating a lower margin of error in the VECM-based predictions. These findings highlight the necessity of data-driven modeling for more reliable cost estimations in the construction industry. The study provides a methodological framework for improving cost forecasting, particularly during periods of high inflation and market uncertainty. By leveraging econometric modeling, decision- makers in the construction sector can enhance financial planning and risk management. Future research may focus on expanding the model with additional economic indicators, testing its applicability in different regions, and developing long-term forecasting strategies to improve predictive accuracy. |
| Keywords: | Construction Cost Estimation, ; Forecasting Models; VECM |
| JEL: | R3 |
| Date: | 2025–01–01 |
| URL: | https://d.repec.org/n?u=RePEc:arz:wpaper:eres2025_232 |
| By: | Elliot Beck; Franziska Eckert; Linus K\"uhne; Helge Liebert; Rina Rosenblatt-Wisch |
| Abstract: | We introduce a novel indicator that combines machine learning and large language models with traditional statistical methods to track sentiment regarding the economic outlook in Swiss news. The indicator is interpretable and timely, and it significantly improves the accuracy of GDP growth forecasts. Our approach is resource-efficient, modular, and offers a way of benefitting from state-of-the-art large language models even if data are proprietary and cannot be stored or analyzed on external infrastructure - a restriction faced by many central banks and public institutions. |
| Date: | 2025–11 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2511.04299 |
| By: | David P. Brown (University of Alberta); Andrew Eckert (University of Alberta); Douglas Silveira (Federal University of Juiz de Fora) |
| Abstract: | The extent of real-time information disclosure in electricity markets has been a longstanding debate. Regulators have the difficult task of striking a careful balance between transparency to improve market outcomes under uncertainty while limiting the potential for coordinated action. We consider the case of Alberta’s electricity market where, until 2017, firms observed anonymized price-quantity offers in the wholesale market in near-real-time. We empirically evaluate the role that this information played in improving firms’ abilities to forecast wholesale prices, a key argument raised by stakeholders for this information to be published. While we find that this information improved firms’ abilities to forecast wholesale prices under certain market conditions, we present evidence to suggest that the economic significance of this improvement is minimal. We point to other types of near-real-time information that could help improve expectations of future market outcomes and provide suggestions on information disclosure policies that aim to strike a balance on motivating efficient outcomes, while reducing the risk of coordination. |
| Keywords: | Machine Learning; Electricity; Price Forecasting; Competition Policy |
| JEL: | D43 L13 L50 L94 Q40 |
| Date: | 2025–10–26 |
| URL: | https://d.repec.org/n?u=RePEc:ris:albaec:021762 |
| By: | Mainak Singha; Jose Aguilera-Toste; Vinayak Lahiri |
| Abstract: | We test whether simple, interpretable state variables-trend and momentum-can generate durable out-of-sample alpha in one of the world's most liquid assets, gold. Using a rolling 10-year training and 6-month testing walk-forward from 2015 to 2025 (2, 793 trading days), we convert a smoothed trend-momentum regime signal into volatility-targeted, friction-aware positions through fractional, impact-adjusted Kelly sizing and ATR-based exits. Out of sample, the strategy delivers a Sharpe ratio of 2.88 and a maximum drawdown of 0.52 percent, net of 0.7 basis-point linear cost and a square-root impact term (gamma = 0.02). A regression on spot-gold returns yields a 43 percent annualized return (CAGR approximately 43 percent) and a 37 percent alpha (Sharpe = 2.88, IR = 2.09) at a 15 percent volatility target with beta approximately 0.03, confirming benchmark-neutral performance. Bootstrap confidence intervals ([2.49, 3.27]) and SPA tests (p = 0.000) confirm statistical significance and robustness to latency, reversal, and cost stress. We conclude that forecast-to-fill engineering-linking transparent signals to executable trades with explicit risk, cost, and impact control-can transform modest predictability into allocator-grade, billion-dollar-scalable alpha. |
| Date: | 2025–11 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2511.08571 |
| By: | Stanislav Selitskiy |
| Abstract: | We investigate a number of Artificial Neural Network architectures (well-known and more ``exotic'') in application to the long-term financial time-series forecasts of indexes on different global markets. The particular area of interest of this research is to examine the correlation of these indexes' behaviour in terms of Machine Learning algorithms cross-training. Would training an algorithm on an index from one global market produce similar or even better accuracy when such a model is applied for predicting another index from a different market? The demonstrated predominately positive answer to this question is another argument in favour of the long-debated Efficient Market Hypothesis of Eugene Fama. |
| Date: | 2025–11 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2511.08658 |