|
on Forecasting |
| By: | Marc Schmitt |
| Abstract: | In algorithmic markets, predictive models become part of the data-generating process they aim to forecast. Once their outputs are converted into trades, allocations, execution schedules, or risk controls, they change the future data on which they are evaluated. I introduce algometrics, a framework for time series whose evolution depends on the predictive algorithms forecasting them. The framework distinguishes historical risk, measured under passive forecasting, from deployment risk, measured when forecasts drive actions. I prove three results. First, deployment risk is not identifiable from passive historical data alone: even in a one-step linear feedback model, infinitely many algorithm-mediated environments induce the same historical law while implying different deployment risks for the same forecaster. Second, historical model rankings can invert under crowding, so a predictor with lower passive error can have higher deployment error once similar algorithms are adopted. Third, randomized or instrumented actions identify short-horizon linear feedback, and I derive a finite-sample bound for deployment-risk estimation. These results suggest that time-series benchmarks in algorithmic markets should report feedback sensitivity alongside predictive accuracy. |
| Date: | 2026–05 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2605.23978 |
| By: | Marie Soehl Coolsaet; Roberto Gallardo; Zhen Gao |
| Abstract: | This research aims to leverage machine learning to improve stock price prediction and support informed investment decisions related to buying, selling, and holding assets. Specifically, this work investigates transformer-based models for stock prediction and examines the impact of pre-training strategies on forecasting performance. A transformer model was first pre-trained on the Toronto Stock Exchange Index (TSX) to predict intra-day return direction and subsequently fine-tuned on individual TSX stocks. The model was further adapted for return-value regression tasks. Performance was benchmarked against Long Short-Term Memory (LSTM) and XGBoost models. Pre-training on the market index improved the binary cross-entropy loss for individual stock prediction from 0.69 to 0.64. The fine-tuned transformer regression model achieved lower mean squared error than the benchmark models, although the ensemble and XGBoost models achieved higher average daily returns. In addition, a practical application was developed to deliver real-time stock predictions for trading support. Future work will focus on increasing transformer model capacity, incorporating broader global technical indicators, and filtering out stocks with low predictability. |
| Date: | 2026–05 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2605.23962 |
| By: | Miguel Sanchez-Martinez (International Labour Organization); Tomasz Wo\'zniak (University of Melbourne) |
| Abstract: | The R package bpvars was designed to forecast employment, unemployment, and labour market participation rates of 189 countries. However, it is generally applicable to dynamic panel data due to the flexibility of its modelling framework and robust coding. It includes a family of Bayesian hierarchical panel Vector Autoregressions (VARs) that are characterised by: (i) country-specific VAR models (ii) with their parameters' priors centred around their global counterparts, and (iii) featuring flexible multi-level hierarchical prior distributions (iv) with many variants of well-established in the literature benchmark choices, and (v) four alternative specifications including groupping of country-specific or global parameters. A~distinguishing feature is its implementation of missing observation treatment based on a model-coherent Bayesian approach. These models are accompanied by Bayesian prediction, offering a wide range of possible specifications that aim to increase forecasting precision and comply with various reporting standards. We also implement pseudo-out-of-sample recursive forecasting for evaluating point and density forecast performance. The package implements model specification, estimation, and forecasting routines, facilitating simple workflows and reproducibility, including estimation and forecasting results summaries and visualisations. It achieves extraordinary computational speed thanks to the employment of frontier econometric and numerical techniques, as well as algorithms written in C++. |
| Date: | 2026–06 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2606.14143 |
| By: | Magdalena Cornejo (Universidad Torcuato Di Tella - CONICET); Walter Sosa Escudero (Universidad de San Andrés - CONICET) |
| Abstract: | This paper studies forecasting in dynamic panel data models with fixed effects. We compare the forecasting accuracy of conventional estimators—pooledOLS, fixed effects, Anderson–Hsiao, and Arellano–Bond—against shrinkage and regularization methods such as Ridge, LASSO, ElasticNet, empirical Bayes maximum likelihood and the recent unbiased risk estimation of Kwon (2026). Monte Carlo evidence shows that shrinkage methods substantially improve out-of-sample accuracy. An empirical application to firm-level leverage dynamics using Compustat data confirms the relevance of these findings for forecasting in corporate finance. Machine learning regularization can improve forecasting performance in dynamic panel settings while preserving the structural framework. |
| Keywords: | Forecasting, Dynamic panel data, Machine learning, Regularization, Corporate finance. |
| JEL: | C53 C58 |
| Date: | 2026–05 |
| URL: | https://d.repec.org/n?u=RePEc:sad:wpaper:183 |
| By: | Olivetti, Leonardo; Messori, Gabriele; Avner, Paolo; Hallegatte, Stephane |
| Abstract: | Assessing the real-world economic value of weather forecasts remains challenging, particularly in the context of high-impact extreme events. Although meteorological skill has improved substantially in recent years—driven by steady advances in physics-based models and impressive breakthroughs in artificial intelligence-based forecasting—operational evaluations still focus primarily on standard skill metrics, with limited consideration of how improvements in meteorological skill translate into economic value. This study proposes a flexible framework to assess the economic value of weather forecasts, with penalty functions that explicitly account for compounding losses as well as declining user trust in cases of repeated false alarms. In addition, the framework allows for varying cost—loss ratios to represent heterogeneous prevention costs and vulnerability structures. The framework is applied to cities exposed to weather-related natural hazards, comparing the relative economic value of leading physics-based and data-driven forecasting systems from the European Centre for Medium-Range Weather Forecasts. The value of forecasts is highly sensitive to assumptions about compounding losses, penalty structures, and prevention costs, which often substantially alter conclusions drawn from meteorological skill alone. For instance, in some cities in Southern Europe, the higher sensitivity of the physics-based Integrated Forecast System high-resolution model (IFS HRES) makes it better suited when protection costs are small relative to potential losses, while the higher specificity of the data-driven Artificial Intelligence Forecasting System (AIFS) makes it better when protection costs are higher. These findings underscore the importance of evaluating economic value under realistic risk scenarios to ensure that improvements in predictive accuracy translate into meaningful societal and economic benefits. |
| Date: | 2026–06–02 |
| URL: | https://d.repec.org/n?u=RePEc:wbk:wbrwps:11407 |
| By: | Degui Li (Faculty of Business Administration, University of Macau); Yuying Sun (Chinese Academy of Sciences); Boyao Wu (University of International Business and Economics) |
| Abstract: | In this paper, we introduce a flexible time-varying multi-layer network vector autoregression (VAR) model framework for large-scale time series, allowing agents in dynamic systems to interact through multiple channels and incorporating multiple adjacency matrices to capture network spillover effects. We propose a penalized model averaging method to determine a time-varying optimal combination of multi-layer network VAR candidate models whose number may be divergent. Under some regularity conditions, the asymptotic properties such as asymptotic optimality and convergence rates of the proposed time-varying weight estimation are derived in the contexts of both the in-sample fitting and out-of-sample prediction. In addition, we extend the conformal prediction method to construct prediction bands for locally stationary time series. Monte-Carlo simulation studies and an empirical application to forecast CPI inflation by combining multiple network information are given to illustrate reliable finite-sample estimation and predictive performance of the developed methodology. |
| Keywords: | asymptotic optimality, conformal prediction, model averaging, multi-layer network, time-varying VAR |
| JEL: | C32 C38 C55 C58 |
| Date: | 2026–06 |
| URL: | https://d.repec.org/n?u=RePEc:boa:wpaper:202640 |
| By: | Manuel Noseda; Nathan Soldati; Marco Paina |
| Abstract: | Predicting stock price movements during Earnings Announcements (EAs) is a significant challenge due to market noise and high-impact price discontinuities. In this study, we evaluate whether pre-announcement news sentiment, firm fundamentals, and recent market dynamics jointly predict the directional price movement of equities on EA days. We construct a multi-modal feature space combining 15 fundamental metrics, 3 price-based technical indicators and sentiment scores derived from financial news articles processed using FinBERT. We compare a Long Short-Term Memory (LSTM) network and a Transformer-based architecture against a logistic regression baseline, and further assess all models with and without sentiment features to quantify their incremental value. Our results indicate that while the LSTM demonstrates higher precision through a conservative safe-bet strategy, the Transformer model exhibits superior sensitivity in identifying volatile movements, achieving a higher macro F1-score, with ablation experiments showing a consistent benefit from incorporating news sentiment. |
| Date: | 2026–05 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2605.25894 |
| By: | Xintong Wu; Peiting Tsai; Jing Yuan; Michael Yu; Greg Sun; Luyao Zhang |
| Abstract: | Decentraland, a decentralized virtual reality platform operating within the expanding Metaverse ecosystem, utilizes its native MANA token to facilitate virtual asset transactions and governance. This study investigates the integration of Discord community sentiment with multi-modal financial data to enhance cryptocurrency price prediction within virtual world economies. We address: (1) identifying sentiment patterns within Decentraland's Discord community, and (2) evaluating the impact of multi-modal features on token return forecasting. Using a BERT-based large language model for sentiment analysis, we develop two LSTM architectures: a baseline incorporating historical prices and a multi-modal variant integrating sentiment scores, trading volume, and market capitalization. Results indicate predominantly neutral community sentiment with a positive skew. The multi-modal model significantly outperforms the price-only baseline in prediction accuracy. These findings demonstrate the predictive value of community-derived signals for virtual economy forecasting and establish a foundation for future research at the intersection of immersive virtual environments, natural language processing, and cryptocurrency market analysis. |
| Date: | 2026–04 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2605.20192 |
| By: | Dennis Shen |
| Abstract: | Estimating causal effects in panel data is a central problem in policy evaluation. Existing methods largely address retrospective questions of the form: what would have happened to a target unit under a different intervention during the observed panel? In many applications, however, decision-makers face prospective questions: what will happen to a target unit under an intervention it has not yet experienced, beyond the observed panel? This article develops a framework for answering such causal forecasting questions by integrating the retrospective counterfactual logic of synthetic-controls-based approaches with the extrapolative structure of multivariate time-series forecasting. Building on the latent factor models that justify unit-side regressions in synthetic controls, we impose low-rank temporal structure on the latent time factors to identify prospective causal forecast estimands. We operationalize this strategy through the Two-Way Synthetic Forecasting estimator, or TWSF, which learns cross-unit relationships from pre-treatment outcomes and combines them with a time-series model learned from treated donor trajectories under the intervention of interest. Under suitable conditions, we establish finite-sample forecasting error bounds that imply pointwise consistency and introduce an orthogonalized correction that yields asymptotic normality and thus enables pointwise inference. We extend the framework to fixed multi-step forecasting horizons through both direct and recursive procedures, each of which inherits analogous pointwise guarantees. We corroborate the theory with simulation studies and illustrate the practical utility of TWSF by studying the public-health impact of opening NFL stadiums during the 2020 season. |
| Date: | 2026–06 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2606.18512 |
| By: | Hwee Kwan Chow (School of Economics, Singapore Management University); Jordan Lee (Singapore Management University) |
| Abstract: | This study empirically assesses the drivers of risks to the inflation outlook for a small open economy like Singapore. We apply the inflation-at-risk framework of López-Salido and Loria (2020) and incorporate projections from the Survey of Professional Forecasters (SPF) as point forecasts of inflation. Our findings show that macro-financial risk factors—shaped by Singapore’s openness, role as a financial hub, and exchange rate–centered monetary policy framework—enter nonlinearly into inflation risk models and exert differentiated effects. Foreign price pressures heighten upside risks, and exchange rate policy has proven effective at mitigating them. Tighter global financial conditions amplify inflation risks through cost-push channels, whereas demand weakness produces only muted downside effects. We also record sharp gains in log predictive scores for one-quarter ahead conditional distributions relative to unconditional ones during the post-pandemic inflation surge. One-year-ahead predictive distributions become markedly right‑skewed ahead of the surge, effectively signalling a heightened probability of extreme inflation outcomes. Overall, incorporating inflation risk measures improves both the in-sample fit and the forecast accuracy of predictive distributions of inflation one and four quarters ahead, offering insights for central banks navigating uncertain global conditions. |
| Keywords: | Inflation-at-risk; survey of professional forecasters; quantile regressions; forecast accuracy |
| JEL: | C21 C53 E31 |
| Date: | 2026–02–01 |
| URL: | https://d.repec.org/n?u=RePEc:ris:smuesw:022912 |
| By: | Sara A. Safari; Christoph Schmidhuber |
| Abstract: | We forecast future volatilities and correlations of financial markets based on the current trends in these markets. This complements previous work that models future expected returns by a cubic polynomial of the current trend strength. Empirically, we observe that volatilities and correlations tend to increase day after day in times of strong up- or down-trends. This effect is particularly pronounced in down-trends. It can be accurately quantified by quadratic polynomials of today's trend strengths, which refine common mean-reversion models of volatilities and correlations. Our results improve the prediction of market risk by accounting for market trends. They also support a recent proposal to model financial markets by a lattice gas near its critical point. |
| Date: | 2026–06 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2606.20145 |
| By: | Ludolph, Melina; Nghiem, Giang; Tonzer, Lena |
| Abstract: | We examine whether combining factual information on inflation levels and forecasts with a narrative can persistently shape consumers' inflation expectations. In a preregistered randomized controlled trial with a representative sample of 3, 000 German consumers, participants received either numerical or textual information about inflation rates, with or without an accompanying narrative. All treatments immediately lower inflation expectations, with numerical information eliciting stronger adjustments. Adding a narrative produces no additional immediate effect, confirming that it conveys no new information. However, only the combination of numerical information with a narrative yields a lasting reduction in inflation expectations and forecast uncertainty still observable after four weeks. Our results suggest that combining precise information with a narrative enhances information retention and can lead to more persistent shifts in consumers' beliefs. The effects are strongest when respondents perceive the narrative as relatable and emotionally engaging, and among those with low financial literacy and limited knowledge of inflation. |
| Keywords: | central bank communication, inflation expectations, narratives |
| JEL: | D84 D91 E31 E58 |
| Date: | 2026 |
| URL: | https://d.repec.org/n?u=RePEc:zbw:iwhdps:341628 |
| By: | Marcel Muller; Arno Botha; Conrad Beyers |
| Abstract: | An integrated and extendable approach for stress-testing loan portfolios is presented, which includes both a loan production component and a credit risk component. In this approach, we simulate a completed portfolio using realistic loan parameters and distributional assumptions. Thereafter, we generate the uncertain cash flow history of these loans within a multistate probabilistic framework. We illustrate our approach using a simulation-based study, though the approach can be fit to real-world data. Such a simulation-based approach is ideal for stress-testing since it allows for evaluating a range of conditions. From these completed loans, we compute portfolio-level credit risk metrics, e.g., default and loss rates. Stress scenarios are introduced by varying the loan parameters accordingly within a broader Monte Carlo setup, thereby resulting in a range of portfolios. A classical approach to stress-testing does not typically integrate loan production or embed the correlation structure amongst risk metrics. In our approach, we integrate the forecasting of risk metrics with receipt-generation. Given data, the loan parameters within our extendable approach can be dynamically modelled as functions of input variables using any applicable technique. Overall, our approach can render predictions that are more dynamic and flexibly tuned, which can enhance stress-testing practices within any bank. |
| Date: | 2026–06 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2606.19052 |