|
on Forecasting |
| By: | Hilde C. Bjornland; Nicolas Hardy; Dimitris Korobilis |
| Abstract: | We develop a Quantile Bayesian Vector Autoregression (QBVAR) to forecast real oil prices across different quantiles of the conditional distribution. The model allows predictor effects to vary across quantiles, capturing asymmetries that standard mean-focused approaches miss. Using monthly data from 1975 to 2025, we document three findings. First, the QBVAR improves median forecasts by 2-5%relative to Bayesian VARs, demonstrating that quantile-specific dynamics matter even for point prediction. Second, uncertainty and financial condition variables strongly predict downside risk, with left-tail forecast improvements of 10-25% that intensify during crisis episodes. Third, right-tail forecasting remains difficult; stochastic volatility models dominate for upside risk, though forecast combinations that include the QBVAR recover these losses. The results show that modeling the conditional distribution yields substantial gains for tail risk assessment, particularly during major oil market disruptions. |
| Keywords: | oil price forecasting, quantile regression, Bayesian VAR, tail risk, distributional forecasting |
| JEL: | C32 C53 Q41 Q47 |
| Date: | 2026–05 |
| URL: | https://d.repec.org/n?u=RePEc:een:camaaa:2026-39 |
| By: | Nam Ho-Nguyen; Hossein Alipour; Anastasios Panagiotelis; George Athanasopoulos |
| Abstract: | Forecasting multivariate data that adhere to known linear constraints, so-called hierarchical data, benefits from a post-processing step known as reconciliation. While traditional reconciliation methods focus on mean forecasts, in a decision-making setting the optimal action is a functional of a belief distribution, for example a quantile. Building on a general framework, this paper develops a new methodology for forecast reconciliation where the objective is to obtain accurate forecasts for a given quantile level. This is achieved by minimising expected pinball loss, a challenging problem which we propose to overcome in two ways. First, expectations are approximated by drawing samples from the base forecasts, making the approach applicable for any distributional form. Second, the pinball loss is approximated with a smooth function, enabling optimisation with first-order methods. Theoretical results are developed proving that the minimiser of the objective function employing these two approximations converges, in the limit, to the minimiser of expected pinball loss. Applications to both simulated and real-world data demonstrate that the proposed methodology delivers statistically significant improvements in forecast accuracy over the widely used MinT benchmark. |
| Keywords: | forecast reconciliation, quantile forecasting, hierarchical time series, pinball loss, probabilistic forecasting, MinT, stochastic gradient descent |
| JEL: | C53 C61 C63 |
| Date: | 2026 |
| URL: | https://d.repec.org/n?u=RePEc:msh:ebswps:2026-4 |
| By: | Théo Metz; Carolina Ulloa-Suárez; Oscar M. Valencia |
| Abstract: | This paper examines the causal effect of Independent Fiscal Institutions (IFIs) on the accuracy of government macroeconomic and fiscal forecasts. Using a novel dataset of real-time budget projections for 55 European and Latin American countries over 1998–2019, we exploit the staggered introduction of IFIs to identify their impact on forecast performance. We find that IFI implementation leads to statistically and economically meaningful reductions in forecast errors for GDP growth, total government revenue, and total government expenditure. Improvements in GDP growth forecasts emerge within two years of implementation, whereas gains in fiscal forecast accuracy materialise more gradually, after four to five years. This dynamic pattern is consistent with reputational and accountability mechanisms, whereby sustained independent scrutiny reshapes governments’ forecasting incentives over time. The results are robust across alternative identification strategies and forecast accuracy measures. Overall, the findings underscore the role of IFIs as institutions that strengthen fiscal credibility by reshaping forecasting incentives over time, thereby complementing fiscal rules and supporting more sustainable fiscal policymaking. |
| Keywords: | Independent Fiscal Institutions; forecast accuracy; fiscal credibility; staggered difference-in-differences; fiscal rules. |
| JEL: | E62 H61 H68 C23 |
| Date: | 2026 |
| URL: | https://d.repec.org/n?u=RePEc:ulp:sbbeta:2026-14 |
| By: | Brooke Hathhorn; Laura E. Jackson; Michael T. Owyang |
| Abstract: | We evaluate the ability of economic uncertainty measures to forecast recessions in real time. We find that including uncertainty increases the predictive power of both in sample and out-of-sample forecast models relative to a baseline set of financial variables. A nonlinear maximum transformation of uncertainty, which captures whether a measure exceeds its maximum over the past year, improves forecast performance for some measures. Adding a contemporaneous indicator like GDP growth alongside uncertainty yields additional predictive gains. Lastly, ex post Bayesian model averaging outperforms individual uncertainty models and ex ante factors of uncertainty generated using principal component analysis. |
| Keywords: | precision-recall curve; receiver-operator characteristic; probit regression; out-of-sample forecasting |
| JEL: | E32 E37 E52 |
| Date: | 2026–05–20 |
| URL: | https://d.repec.org/n?u=RePEc:fip:fedlwp:103289 |
| By: | Thomas K. Kloster; Fred Espen Benth |
| Abstract: | We study forecasting of the realized covariation in electricity markets. The realized covariation in this context is a matrix-valued representation of the latent infinite-dimensional covariance operator and a parsimonious matrix-HAR type model is constructed to facilitate estimation. We test the model on one-week ahead forecasts of the weekly realized covariation and find that the inclusion of longer time horizons and renewable generation information adds important predictive power. We also investigate the prediction of risk premia in electricity forward markets and find that our variance forecasts provide substantially improved forecasts of spread risk premia compared to standard methods relying on backward looking volatility. |
| Date: | 2026–06 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2606.05991 |
| By: | Tanisa Tawichsri; Suppawong Tuarob; Nuwat Nookhwun; Chinjuta Sangasaeng |
| Abstract: | We develop a news-based inflation expectations index for Thailand using a scalable workflow that integrates topic modeling, LLM-assisted labeling, and fine-tuned BERT classification. Based on 1.1 million Thai-language news articles from 2015–2024, the index leads both headline inflation and firm inflation expectations. Given that inflation narratives in news are inherently subjective and often ambiguous, we show that prompt design can materially affect downstream economic inference. In out-ofsample forecasting, augmenting autoregressive benchmarks with the news index reduces RMSE by up to 32% for headline inflation and 30% for firm inflation expectations, with gains increasing at longer horizons. SHAP-based decomposition reveals a horizon-dependent information structure: price-specific topics drive short-term forecasts, while macroeconomic narratives dominate at longer horizons. Our findings demonstrate that LLM-assisted text analysis can generate economically meaningful inflation indicators in non-English, emerging-economy settings. The index also performs particularly strong during periods of elevated inflation uncertainty. |
| Keywords: | Inflation expectations; Text-based indicators; Online news data; Large language models (LLMs); Machine learning; Sentiment analysis; Nowcasting and forecasting; Emerging economies |
| JEL: | E31 E37 D84 |
| Date: | 2026–05 |
| URL: | https://d.repec.org/n?u=RePEc:pui:dpaper:252 |
| By: | Labastidas, Esteban |
| Abstract: | We develop and evaluate a hybrid early-warning system for year-over-year (YoY) inflation in Colombia that combines four econometric models (ARIMA, LASSO, ElasticNet, and a weighted ensemble), a reduced-form VAR, an agent-based model with heterogeneous expectations and tradable/non-tradable pass-through (ABM v2), a large language model (LLM) forecaster, and a multi-output agent architecture, integrated through Dynamic Model Averaging (DMA). We evaluate the system on a rolling out-of-sample backtest from February 2010 to March 2026 (n ≈ 194 months) spanning the 2021–2023 inflation surge and its ongoing disinflation. Five contributions emerge. First, an identity-based monthly-to-YoY decomposition applied uniformly across reduced-form models reduces MAE by 15–20% relative to direct YoY forecasting without adding variables. Second, regime-conditional analysis shows that MAE is 2.0–3.0 times larger in surge regimes than in stable regimes across all models. Third, an ABM with regime-dependent heterogeneous expectations reduces full-sample MAE from 0.337 pp (v1) to 0.268 pp (v2, −20.4%) and surge-regime MAE from 0.585 pp to 0.404 pp (−31.0%), with Diebold-Mariano statistic 6.21 (p |
| Keywords: | Inflation forecasting, Emerging markets, Agent-based models, Heterogeneous expectations, Large language models, Dynamic model averaging, Diebold-Mariano test, Colombia |
| JEL: | C45 C53 E31 E37 |
| Date: | 2026–04–18 |
| URL: | https://d.repec.org/n?u=RePEc:pra:mprapa:128779 |
| By: | Jayesh Chaudhary |
| Abstract: | Financial volatility exhibits substantial non-stationarity, making single-regime models inadequate for characterising changing market conditions. This paper proposes a triple-timeframe Markov-Switching GARCH (MS-GARCH) framework for volatility regime detection in EUR/USD across daily, four-hour, and hourly horizons. Three independent AR(1)-MS-GARCH models are estimated to capture macro, meso, and micro regime dynamics, while Filardo-style time-varying transition probabilities (TVTP) are incorporated at the shorter horizons through composite stress indicators. The resulting regime probabilities are combined through an outer-product construction into a 27-state cross-scale probability tensor. Using EUR/USD data from 2015-2025, the framework produces statistically distinct Calm, Turbulent, and Crisis regimes and achieves superior out-of-sample volatility forecasting performance relative to a conventional GARCH benchmark. The results suggest that volatility dynamics contain meaningful structure across multiple timescales and that modelling these scales separately provides a more informative representation of market conditions than a single-timescale approach. |
| Date: | 2026–06 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2606.06190 |
| By: | Rylan Wade |
| Abstract: | This paper tests whether graph neural networks improve realized volatility forecasts and whether those forecasts improve portfolio performance. Using weekly realized volatility for 465 S\&P 500 equities from 2015--2025, Heterogeneous Autoregressive and Long Short-Term Memory baselines are compared against GraphSAGE models built on rolling correlation, sector, and Granger-causal graphs, with and without macro regime features. The empirical finding is that the model with the lowest forecast MSE, the model with the highest cross-sectional ranking accuracy, and the model with the highest portfolio Sharpe ratio are three different models. Forecast accuracy, ranking quality, and portfolio performance are related but not interchangeable objectives. Graph volatility models add value only when the portfolio rule can exploit the cross-sectional structure they encode. |
| Date: | 2026–05 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2605.19278 |
| By: | Eric Engstrom |
| Abstract: | In January 2012, the Federal Reserve began publishing the Summary of Economic Projections (SEP) "dot plot, " revealing FOMC participants' projections for the federal funds rate. This paper documents a dual role for SEP projections in the formation of private interest-rate expectations. On one hand, SEP projections contain valuable information, achieving lower forecast errors than consensus surveys, VAR models, and several market-based measures at many horizons. Because the SEP is informative, some reliance on it by private forecasters is natural. On the other hand, because the SEP is updated only quarterly, SEP projections that are useful when released can become stale between updates. If private forecasts continue to place excessive weight on those earlier projections, they may respond too slowly to newly arriving information. Consistent with this prediction, survey forecast errors-and, to a weaker extent, market-based forecast errors-are systematically related to the gap between current expectations and lagged SEP projections, even after controlling for macroeconomic conditions, risk premia, and other predictors of forecast errors. The findings imply that official guidance can simultaneously improve average forecast accuracy while reducing the speed with which new information is incorporated into expectations. |
| Keywords: | anchoring bias; monetary policy expectations; Federal Reserve communications; forecast efficiency; dot plot; term structure |
| JEL: | E43 E47 E52 E58 G12 G14 |
| Date: | 2026–05–14 |
| URL: | https://d.repec.org/n?u=RePEc:fip:fedgfe:103336 |
| By: | Nobuyuki Hanaki; Bolin Mao; Tiffany Tsz Kwan Tse; Wenxin Zhou |
| Abstract: | This study investigates willingness to pay (WTP) for stock forecasting advice from algorithms, financial experts, and peers. In two incentivized forecasting experiments, participants purchased advice using an incentive-compatible mechanism and then decided how much to incorporate it into their forecasts. Participants assigned the highest WTP to algorithmic advice and relied on it as heavily as expert advice, despite its forecasting performance being no better than alternative sources. Consequently, participants overpaid for advice, especially algorithmic advice, whose realized benefits were insufficient to offset its cost. A second experiment shows that overpayment persists even after repeated opportunities to revise WTP with detailed feedback on advice quality and realized net benefits. The results suggest that individuals place excessive value on algorithmic advice perceived as sophisticated or credible, even when its realized economic value is limited. These findings highlight the importance of tools and disclosure policies that help individuals better assess the economic value of algorithmic advice. |
| Date: | 2024–12 |
| URL: | https://d.repec.org/n?u=RePEc:dpr:wpaper:1268rr |
| By: | Sai Ma |
| Abstract: | Using microdata from the Michigan Survey of Consumers, we study how within-household reallocations of attention across news affect inflation expectation bias, measured relative to a real-time, machine-learning full-information benchmark. Shifting attention toward unfavorable (favorable) economic news increases (decreases) forecast bias substantially, while dropping attention to an unfavorable topic has little effect. The largest bias increases come not from inflation news itself, but from attention to unfavorable social, political, and geopolitical narratives. Aggregate news sentiment has no effect on bias when a household's reported attention allocation is unchanged. In aggregate, these effects are amplified when the attention network is dominated by an unfavorable focal hub: bias-reducing favorable narratives are crowded out of limited attention sets, and respondents closer to the hub exhibit larger bias increases. We find that past and present attention to news together account for up to 70 percent of observed forecast bias, with the current attention component rising sharply during recessions and large negative news events. Results are robust to a battery of specification checks and external validation. |
| Keywords: | inflation expectations; limited attention; forecast bias; sentiment; networks |
| JEL: | E31 E52 D83 D84 |
| Date: | 2026–04–17 |
| URL: | https://d.repec.org/n?u=RePEc:fip:fedgif:103344 |
| By: | Mamiko Yamashita (Osaka School of International Public Policy, the University of Osaka) |
| Abstract: | This paper studies the relationship between option-implied, risk-neutral forecasts and their real-world counterparts through the lens of stochastic dominance and pricing kernel monotonicity. We show that when the pricing kernel is weakly decreasing in asset payoffs, the real-world distribution first-order stochastically dominates the risk-neutral one, implying that the risk-neutral forecast is more conservative. The ordering is reversed when the pricing kernel is weakly increasing, implying that risk-neutral forecasts may be more optimistic rather than conservative. We further show that this monotonicity is closely linked to the dependence between asset payoffs and aggregate consumption. Our results provide a new perspective on the pricing kernel puzzle, an empirical finding that pricing kernels for major market indices are often non-monotonic. Our results, together with the pricing kernel puzzle, suggest that the commonly held belief in the conservativeness of risk-neutral forecasts is not generally warranted, even for broad market indices. |
| Keywords: | Asset Pricing, Risk Management, Pricing Kernel Puzzle, Stochastic Dominance |
| JEL: | G12 G13 G32 |
| Date: | 2026–05 |
| URL: | https://d.repec.org/n?u=RePEc:osp:wpaper:26e008 |