|
on Forecasting |
| By: | Rahul Billakanti; Minchul Shin |
| Abstract: | We propose a simple binarization of predictors, an "at-risk" transformation, as an alternative to the standard practice of using continuous, standardized variables in recession forecasting models. By converting predictors into indicators of unusually weak states based on a thresholding rule estimated from training data, we demonstrate their ability to capture the discrete nature of rare events such as U.S. recessions. Using a large panel of monthly U.S. macroeconomic and financial data, we show that binarized predictors consistently improve out-of-sample forecasting performance, often making linear models competitive with flexible machine learning methods, and that the gains are particularly pronounced around the onset of recessions. |
| Date: | 2026–03 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2603.07813 |
| By: | Harris Buchanan; Eric Benhamou |
| Abstract: | Trend-following strategies underpin many systematic trading approaches yet struggle under nonstationary and nonlinear market regimes. We propose an LSTM-based framework to forecast next-day trend differences ($\Delta_t$) for the top 30 S\&P 500 equities, validated across market cycles (2005--2025). Key contributions include: (i) formal proof of bias-variance reduction via differencing, (ii) exhaustive empirical benchmarks against OLS, Ridge, and Lasso, (iii) portfolio simulations confirming economic gains in terms of overall PNL compared to other models like OLS, Ridge, Lasso or LightGBM Regressor |
| Date: | 2026–03 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2603.14453 |
| By: | Minchul Shin; Nathan Schor |
| Abstract: | We introduce ForeComp, an R package for comparing predictive accuracy using Diebold-Mariano type tests of equal predictive ability with standard and fixed smoothing inference. The package provides a common interface for loss differential based testing and includes Plot Tradeoff, a visual diagnostic for bandwidth sensitivity and the size-power tradeoff. We illustrate the toolkit with Survey of Professional Forecasters applications and Monte Carlo evidence on finite-sample performance. |
| Date: | 2026–03 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2603.07458 |
| By: | Pei-Jun Liao; Hung-Shin Lee; Yao-Fei Cheng; Li-Wei Chen; Hung-yi Lee; Hsin-Min Wang |
| Abstract: | Predicting stock prices presents challenges in financial forecasting. While traditional approaches such as ARIMA and RNNs are prevalent, recent developments in Large Language Models (LLMs) offer alternative methodologies. This paper introduces an approach that integrates LLMs with daily financial news for stock price prediction. To address the challenge of processing news data and identifying relevant content, we utilize stock name embeddings within attention mechanisms. Specifically, we encode news articles using a pre-trained LLM and implement three attention-based pooling techniques -- self-attentive, cross-attentive, and position-aware self-attentive pooling -- to filter news based on stock relevance. The filtered news embeddings, combined with historical stock prices, serve as inputs to the prediction model. Unlike prior studies that focus on individual stocks, our method trains a single generalized model applicable across multiple stocks. Experimental results demonstrate a 7.11% reduction in Mean Absolute Error (MAE) compared to the baseline, indicating the utility of stock name embeddings for news filtering and price forecasting within a generalized framework. |
| Date: | 2026–03 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2603.19286 |
| By: | Keonvin Park |
| Abstract: | Portfolio construction traditionally relies on separately estimating expected returns and covariance matrices using historical statistics, often leading to suboptimal allocation under time-varying market conditions. This paper proposes a joint return and risk modeling framework based on deep neural networks that enables end-to-end learning of dynamic expected returns and risk structures from sequential financial data. Using daily data from ten large-cap US equities spanning 2010 to 2024, the proposed model is evaluated across return prediction, risk estimation, and portfolio-level performance. Out-of-sample results during 2020 to 2024 show that the deep forecasting model achieves competitive predictive accuracy (RMSE = 0.0264) with economically meaningful directional accuracy (51.9%). More importantly, the learned representation effectively captures volatility clustering and regime shifts. When integrated into portfolio optimization, the proposed Neural Portfolio strategy achieves an annual return of 36.4% and a Sharpe ratio of 0.91, outperforming equal weight and historical mean-variance benchmarks in terms of risk-adjusted performance. These findings demonstrate that jointly modeling return and covariance dynamics can provide consistent improvements over traditional allocation approaches. The framework offers a scalable and practical alternative for data-driven portfolio construction under nonstationary market conditions. |
| Date: | 2026–03 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2603.19288 |
| By: | Martin Jaraiz |
| Abstract: | How much macroeconomic information is contained in a single input-output table? We feed FIGARO 64-sector symmetric tables into DEPLOYERS, a Darwinian agent-based simulator, producing genuine out-of-sample GDP forecasts. For each year, the model reads one FIGARO table for year N, self-organizes an artificial economy through evolutionary selection, then runs 12 months of autonomous free-market dynamics whose emergent growth rate predicts year N+1. The I-O table is the only input: no time series, no estimated parameters, no expectations formation, no external forecasts. We present five results. First, a 9-year Austrian panel (2010-2018) using 12-seed ensembles produces MAE of 1.22 pp overall; for five non-crisis years, MAE falls to 0.42 pp -- comparable to the best professional forecaster (WIFO: 0.48 pp). Second, cross-country portability is demonstrated across multiple FIGARO countries with zero parameter changes. Third, a German 9-year panel reveals systematic +3.7 pp positive bias from export dependency -- an informative negative result. Fourth, a COVID-19 simulation demonstrates the I-O structure as a shock propagation mechanism: a 19-month timeline produces Year 1 GDP -4.62% vs empirical -6.6%. Fifth, emergent firm size distributions match European Commission data without micro-target calibration. These results establish the I-O table as serving a dual purpose: structural baseline engine and dynamic shock propagation mechanism. Since FIGARO covers 46 countries, the approach is immediately portable without retuning parameters. |
| Date: | 2026–03 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2603.12412 |
| By: | Mohammad Al Ridhawi; Mahtab Haj Ali; Hussein Al Osman |
| Abstract: | Stock markets exhibit regime-dependent behavior where prediction models optimized for stable conditions often fail during volatile periods. Existing approaches typically treat all market states uniformly or require manual regime labeling, which is expensive and quickly becomes stale as market dynamics evolve. This paper introduces an adaptive prediction framework that adaptively identifies deviations from normal market conditions and routes data through specialized prediction pathways. The architecture consists of three components: (1) an autoencoder trained on normal market conditions that identifies anomalous regimes through reconstruction error, (2) dual node transformer networks specialized for stable and event-driven market conditions respectively, and (3) a Soft Actor-Critic reinforcement learning controller that adaptively tunes the regime detection threshold and pathway blending weights based on prediction performance feedback. The reinforcement learning component enables the system to learn adaptive regime boundaries, defining anomalies as market states where standard prediction approaches fail. Experiments on 20 S&P 500 stocks spanning 1982 to 2025 demonstrate that the proposed framework achieves 0.68% MAPE for one-day predictions without the reinforcement controller and 0.59% MAPE with the full adaptive system, compared to 0.80% for the baseline integrated node transformer. Directional accuracy reaches 72% with the complete framework. The system maintains robust performance during high-volatility periods, with MAPE below 0.85% when baseline models exceed 1.5%. Ablation studies confirm that each component contributes meaningfully: autoencoder routing accounts for 36% relative MAPE degradation upon removal, followed by the SAC controller at 15% and the dual-path architecture at 7%. |
| Date: | 2026–03 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2603.19136 |
| By: | Germain, Arnaud (Université catholique de Louvain, LIDAM/ISBA, Belgium); Vrins, Frédéric (Université catholique de Louvain, LIDAM/LFIN, Belgium) |
| Abstract: | We introduce a new forecast combination strategy called clagging (for cluster aggregating), which consists in combining models fitted on different clusters. First, we perform K clustering tasks of the same training set, increasing the number of clusters from 1 to K. Next, we fit a model on each of those 1 + 2 +. . . + K clusters. Finally, the aggregate forecast for a new observation is obtained by combining the forecasts of the corresponding models using the distance of the new observation to the clusters’ centroids. We perform an extensive horse race study where we benchmark clagging on 20 datasets using 7 prediction models, considering both regression and classification tasks. Our results suggest that clagging outperforms bagging, where a bootstrapped sample is traditionally created by drawing observations with replacement until the size of the bootstrapped sample coincides with the size of the original training set. Clagging also improve the performance compared to a standard fit on the whole training set. |
| Date: | 2026–03–09 |
| URL: | https://d.repec.org/n?u=RePEc:ajf:louvlf:2026002 |
| By: | Timo Dimitriadis; Marius Puke |
| Abstract: | We introduce inference methods for score decompositions, which partition scoring functions for predictive assessment into three interpretable components: miscalibration, discrimination, and uncertainty. Our estimation and inference relies on a linear recalibration of the forecasts, which is applicable to general multi-step ahead point forecasts such as means and quantiles due to its validity for both smooth and non-smooth scoring functions. This approach ensures desirable finite-sample properties, enables asymptotic inference, and establishes a direct connection to the classical Mincer-Zarnowitz regression. The resulting inference framework facilitates tests for equal forecast calibration or discrimination, which yield three key advantages. They enhance the information content of predictive ability tests by decomposing scores, deliver higher statistical power in certain scenarios, and formally connect scoring-function-based evaluation to traditional calibration tests, such as financial backtests. Applications demonstrate the method's utility. We find that for survey inflation forecasts, discrimination abilities can differ significantly even when overall predictive ability does not. In an application to financial risk models, our tests provide deeper insights into the calibration and information content of volatility and Value-at-Risk forecasts. By disentangling forecast accuracy from backtest performance, the method exposes critical shortcomings in current banking regulation. |
| Date: | 2026–03 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2603.04275 |
| By: | Xiaochun Liu; Richard Luger |
| Abstract: | We introduce a semiparametric approach for forecasting Value-at-Risk (VaR) and Expected Shortfall (ES) by modeling the conditional scale of financial returns, defined as the difference between two specified quantiles, via restricted quantile regression. Focusing on downside risk, VaR is derived from the left-tail quantile of rescaled returns, and ES is approximated by averaging quantiles below the VaR level. The method delivers robust, distribution-free estimates of extreme losses and captures skewness, heavy tails, and leverage effects. Simulation experiments and empirical analysis show that it often outperforms established models, including GARCH and joint VaR-ES conditional-quantile approaches. An application to daily returns on major international stock indices, spanning the COVID-19 period, highlights its effectiveness in capturing risk dynamics. |
| Date: | 2026–03 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2603.02357 |
| By: | Fatemeh Keivani (Universitat de Girona); Germ\`a Coenders (Universitat de Girona); Ge\`orgia Escaram\'is (Universitat de Girona) |
| Abstract: | Using standard financial ratios as variables in statistical analyses has been related to several serious problems, such as extreme outliers, asymmetry, non-normality, and non-linearity. The compositional-data methodology has been successfully applied to solve these problems and has always yielded substantially different results when compared to standard financial ratios. An under-researched area is the use of financial log-ratios computed with the compositional-data methodology to predict bankruptcy or the related terms of business default, insolvency or failure. Another under-researched area is the use of machine learning methods in combination with compositional log-ratios. The present article adapts the classical Altman bankruptcy prediction model and some of its extensions to the compositional methodology with pairwise log-ratios and three common statistical and machine learning tools: logistic regression models, k-nearest neighbours, and random forests, and compares the results with standard financial ratios. Data from the sector in the Spanish economy with the largest number of bankrupt firms according to the first two digits of the NACE code (46XX "wholesale trade, except of motor vehicles and motorcycles") were obtained from the Iberian Balance sheet Analysis System. The sample size (31, 131 firms, of which 97 were bankrupt) was divided into a training and a validation dataset. The training data set was downsampled to one healthy firm to each bankrupt firm. No outliers were removed. Focusing on predictive performance, the results show that compositional methods are better than standard ratios in terms of sensitivity, with mixed results regarding specificity, compositional random forests and compositional logistic regression behaving the best. |
| Date: | 2026–03 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2603.24215 |
| 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); 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 introduce credit standards from the Federal Reserve's Senior Loan Officer Opinion Survey (SLOOS) as a novel predictor of U.S. stock market realized volatility over 1990:04-2024:12. We show that tighter credit standards significantly predict higher realized volatility both in- and out-of-sample at one-, three-, and six-month-ahead horizons. A parsimonious model with only the credit standards factor outperforms more complex specifications incorporating macroeconomic factors, uncertainty indexes, and realized moments, estimated via elastic-net and random forest methods, with forecasting gains increasing at longer horizons. These findings establish credit standards as a powerful and distinct predictor of stock market volatility with practical implications for portfolio allocation and risk management. |
| Keywords: | Credit conditions, Realized stock market volatility, Forecasting |
| JEL: | C22 C53 E23 G10 G17 |
| Date: | 2026–03 |
| URL: | https://d.repec.org/n?u=RePEc:pre:wpaper:202607 |
| By: | Jonathan Benchimol; Sathya Mellina |
| Abstract: | We study how inflation-related language in Federal Reserve communication reprices market-based inflation compensation along the term structure. Using a domain-adapted transformer, we extract stance and inflation-narrative indices from post-meeting statements and Chair press conferences and embed them in a two-layer event study that conditions on target-rate surprises. We trace responses in breakeven inflation (BEI) yields and forwards from two to ten years and complement daily estimates with intraday BEI changes in narrow announcement windows. Four findings emerge. First, statement inflation language lowers BEI compensation across maturities, consistent with markets interpreting the committee-vetted document through the policy reaction function; this pattern is driven most strongly by the Delphic inflation index and is robust to alternative identification. Second, press conferences display the opposite pattern: inflation narratives are associated with positive repricing of long-horizon BEI forwards, consistent with the Chair conveying incremental information about medium-run inflation risks beyond the statement. Third, within press conferences, Delphic language maps into long-horizon forward repricing, while Odyssean language compresses belly-of-curve forwards, consistent with stabilization-window mechanisms. Fourth, intraday evidence shows a sign reversal: communication indices raise BEI compensation during the press conference, whereas the statement window drives the negative daily effect, clarifying how the two communication objects combine to produce net announcement-day repricing. Overall, inflation compensation is not monolithic: its repricing depends on communication format, narrative content, intraday timing, and maturity. |
| Keywords: | central bank communication, inflation compensation, high-frequency identification, event-study methods, Delphic and Odyssean forward guidance, natural language processing |
| JEL: | C45 E43 E52 E58 G12 G14 |
| Date: | 2026–03 |
| URL: | https://d.repec.org/n?u=RePEc:een:camaaa:2026-21 |
| By: | Dardati, Evangelina; Laurent, Thibault; Margaretic, Paula; Paredes, Ean; Thomas-Agnan, Christine |
| Abstract: | This paper evaluates the role of climate variables in predicting international migration by proposing two alternative modeling approaches: scalar-on-composition and scalar-on-density regressions. We compare them with the standard scalar-on-scalar approach. Although most studies rely on annual averages of daily temperatures, focusing solely on central measures can mask essential details, such as nonlinearities and threshold effects. Using the full temperature distribution, either by binning or smoothing, the proposed models achieve improved predictive performance out-of-sample. These gains highlight the importance of properly handling the compositional nature of daily temperature bin data to avoid misleading interpretation of the estimates and flawed inferences. Finally, we demonstrate how incorporating complete temperature distributions into alternative climate scenarios can substantially affect projected outmigration. |
| Keywords: | compositional data; temperature; migration projections; climate change |
| JEL: | C25 C46 Q54 |
| Date: | 2026–03 |
| URL: | https://d.repec.org/n?u=RePEc:tse:wpaper:131610 |
| By: | Federico Vittorio Cortesi; Giuseppe Iannone; Giulia Crippa; Tomaso Poggio; Pierfrancesco Beneventano |
| Abstract: | Neural networks applied to financial time series operate in a regime of underspecification, where model predictors achieve indistinguishable out-of-sample error. Using large-scale volatility forecasting for S$\&$P 500 stocks, we show that different model-training-pipeline pairs with identical test loss learn qualitatively different functions. Across architectures, predictive accuracy remains unchanged, yet optimizer choice reshapes non-linear response profiles and temporal dependence differently. These divergences have material consequences for decisions: volatility-ranked portfolios trace a near-vertical Sharpe-turnover frontier, with nearly $3\times$ turnover dispersion at comparable Sharpe ratios. We conclude that in underspecified settings, optimization acts as a consequential source of inductive bias, thus model evaluation should extend beyond scalar loss to encompass functional and decision-level implications. |
| Date: | 2026–03 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2603.02620 |