|
on Forecasting |
| By: | Maya Kodeih; Aliaa Alnaggar; Mucahit Cevik |
| Abstract: | Monetary-policy announcements and central-bank communications play a central role in foreign exchange markets, yet their qualitative, unstructured form makes their forecasting value difficult to quantify. While prior research has largely focused on sentiment extracted from financial news, comparatively little is known about the relative contribution of different dimensions of monetary-policy communication. Existing studies primarily evaluate whether textual information improves overall forecasting performance but provide limited insight into which communication channels drive such improvements. To address this gap, this paper introduces a statistical attribution methodology that decomposes monetary-policy communication into interpretable channels and quantifies their incremental forecasting contribution under false-discovery-rate control. Monetary-policy news is transformed into structured communication signals using large language models (LLMs) and temporal feature engineering. These signals are evaluated using rolling-window experiments with tree-based machine-learning models. The results show that monetary-policy communication contains measurable predictive information. Attribution analysis shows that predictive value is concentrated in a small subset of signals, with communication timing providing the strongest individual feature-level contribution, targeted communication-activity measures also contributing positively, and LLM-derived sentiment providing complementary information at the group level. The findings indicate that communication-based forecasting value extends beyond sentiment alone and that attribution, rather than aggregate accuracy alone, is central to evaluating news-derived signals. |
| Date: | 2026–09 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2609.32773 |
| By: | Freddy García-Albán; Manuel González-Astudillo |
| Abstract: | This paper investigates which features of a Bayesian dynamic factor model improve U.S. GDP nowcasts and whether their contribution varies across historical episodes. Using pseudo-real-time information sets for 25 series, we estimate 16 combinations of dynamic heterogeneity, stochastic volatility, time-varying long-run growth, and a multiplicative outlier adjustment. We evaluate their marginal contributions to point, density, and quantile score forecasts using balanced paired factorial contrasts. Dynamic heterogeneity improves point, density, and tail forecasts, and it is the only feature that helps consistently across all three criteria. Stochastic volatility improves density and tail forecasts in some periods but not point forecasts consistently. Time-varying long-run growth lowers squared-error loss in some periods but worsens tail forecasts. The outlier adjustment does not improve point or density forecasts overall, has favorable but imprecisely estimated effects on upper-tail quantile scores during the pandemic, and worsens those scores afterward. |
| Keywords: | Bayesian dynamic factor models; forecast evaluation; stochastic volatility; longrun growth; outliers |
| JEL: | C11 C32 C53 E37 |
| Date: | 2026–10–02 |
| URL: | https://d.repec.org/n?u=RePEc:fip:fedgfe:103878 |
| By: | Tae-Hwy Lee (Department of Economics, University of California Riverside); Daanish Padha (Adam Smith Business School, University of Glasgow) |
| Abstract: | We extend the Three-Pass Regression Filter (3PRF) in two dimensions. First, we allow the factors driving the predictor panel to be weak. Second, we allow the idiosyncratic components of the predictors to contain additional predictive information for the forecasting target. We establish conditions under which 3PRF consistently recovers the target-relevant factor space under weak factor structures and characterize how its convergence rate depends on the relative strength of relevant and irrelevant factors. Stronger relevant factors improve the rate at which 3PRF approaches the infeasible factor forecast, whereas stronger irrelevant factors have the opposite effect. We use these results to compare 3PRF with Principal Component Regression (PCR) and identify settings in which either procedure has a relative advantage. To accommodate predictive information outside the common factor space, we augment 3PRF with a Least Absolute Shrinkage and Selection Operator (LASSO) regression on estimated idiosyncratic components, yielding the 3PRF LASSO estimator. We derive its prediction-error rate while accounting for the fact that the idiosyncratic regressors are generated from a first-stage factor estimator. Monte Carlo simulations support the asymptotic results and indicate strong finite-sample forecasting performance of 3PRF LASSO. In an empirical application involving forecasts of U.S. headline CPI inflation using FRED-QD data, 3PRF LASSO delivers competitive forecast performance. |
| Keywords: | Weak Factors; Forecasting; High-dimensional data; Supervision; Three pass regression filter; LASSO |
| JEL: | C18 C22 C53 C55 E27 |
| Date: | 2026–10 |
| URL: | https://d.repec.org/n?u=RePEc:ucr:wpaper:202607 |
| By: | Lenza, Michele; Marchini, Claudia |
| Abstract: | This study applies a Local Linear Forest (LLF) for wage forecasting in France. The LLF outperforms benchmarks such as Random Walk (RW), Ridge Regressions (RR) and Random Forests (RF). We also show that adding foreign predictors (i.e. measures of real activity, price and wage pressures from Germany and Italy) to French economic variables significantly improves wage predictions in France. The implications of our results are that wage dynamics in France exhibit non-linearity, to a certain extent, and that the better ability of LLFs relative to RFs to fit smooth signals is a valuable feature, especially at times in which wage growth reaches unprecedented levels from an historical perspective. JEL Classification: C45, C55, E37, J30 |
| Keywords: | forecasting, local linear forest, negotiated wages |
| Date: | 2026–10 |
| URL: | https://d.repec.org/n?u=RePEc:ecb:ecbwps:20263291 |
| By: | David Schaurecker; Lasse B. Strand; Kevin O'Sullivan; Robert Jakob |
| Abstract: | Short-horizon price-trend prediction from limit order books in equity and intraday electricity markets requires models that combine predictive quality with low single-sample latency and a small serialized model size to keep pace with rapid and continuous market updates. We introduce MBOFormer, a 7, 203-parameter causal transformer, and MBOFusion, a 14, 371-parameter extension with a slow temporal-context branch. Both models process market-by-order (level-3) histories of individual order submissions, cancellations, and executions. We compare them with level-2-based baselines that process sampled order-book snapshots and simple statistics instead of their underlying messages. Across three markets and four prediction horizons, our models achieve the highest mean macro-F1 in eleven of the twelve settings in a comparison against other benchmark models of varying sizes. Both our models achieve sub-millisecond median inference on a single Apple M2 CPU thread and have serialized state dictionaries below 70 kB. MBOFormer is 5.1x-8.1x faster than the million-parameter baselines at comparable predictive quality. Our results show that fine-grained market event history can reduce the need for model size under tight deployment constraints for short-term equity and electricity price forecasting tasks. |
| Date: | 2026–10 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2610.02917 |
| By: | Feras A. Saad; Todd B. Walker |
| Abstract: | Seasonal adjustment is fundamental to economic analysis, but uncertain because seasonal components are inherently latent. This article introduces a probabilistic model discovery method that decomposes a time series into seasonal and nonseasonal components. The method returns a posterior distribution over the structure and parameters of a seasonal component. In simulation studies, the method can improve point forecasts, interval predictions, and recovery of seasonal components relative to X-13ARIMA-SEATS. In a study of eight U.S. macroeconomic series during the COVID-19 recession, the method surfaces significant ex-ante uncertainty about current seasonal adjustments in real time, well before many X-13 revisions reach their eventual peaks. |
| Date: | 2026–09 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2609.36280 |
| By: | Anna B. Schmidt |
| Abstract: | This work studies how Bayesian machine learning methods can be used for large-scale demand estimation with many product categories. I compare two model classes, a latent factorization model and a mixed logit model and two Bayesian estimation approaches, Markov Chain Monte Carlo (MCMC) and Variational Inference (VI). The analysis combines a simulation study with an application to supermarket scanner data. The results show that the latent factorization model benefits from information across categories and improves its predictive performance as the dimensionality of the choice environment increases, whereas the mixed logit model does not exhibit the same pattern. MCMC delivers the highest predictive accuracy but is computationally intensive. VI achieves slightly lower predictive performance while substantially reducing runtime. In the empirical application, VI also outperforms the mixed logit benchmark. These findings highlight a trade-off between accuracy and computational feasibility in multi-category demand estimation. |
| Date: | 2026–10 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2610.08409 |
| By: | Md Rezwanul Islam; Wael Mohammed |
| Abstract: | Global forecasting models pool many series and learn one shared function. Gradient-boosted trees are their most common form. We measure a failure of this design that has not, to our knowledge, been documented. Train a global tree on the individual series of a hierarchy, then ask it for the hierarchical aggregate. The aggregate sits far outside the model's training range, and the forecast collapses. The model under-predicts the total by 30-50x in our production deployment, and by up to 496x in a public M5 reconstruction. The mechanism is known: beyond its training range, a tree predicts a constant. It surfaces at the aggregate because the total dwarfs every training series. The cure is not new. Per-series scaling, the preprocessing step that Montero-Manso and Hyndman (2021) recommend, prevents the collapse. So do a weighted aggregate-level training row and seasonal differencing. Our contribution is the characterization. The collapse reproduces on five panels: a production business-to-business marketplace, a synthetic hierarchy, M5, Australian Tourism, and a public business-buyer panel. It holds on three tree libraries, is invariant across training seeds, and is statistically significant. Its onset is immediate and tracks a simple support bound: a scale gap of only 1.15x already costs a third of the total. No standard configuration change prevents it: pooling every hierarchy level into training fails at scale, and the one knob that fits linear models in the leaves softens it without curing it. Rolling the forecasts forward recursively separates the cures: the aggregate-row cure re-collapses, per-series scaling degrades but stays low, and only seasonal differencing keeps its one-step accuracy unchanged. We close with a three-step procedure for diagnosing and preventing the failure in deployed systems. |
| Date: | 2026–08 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2609.27912 |
| By: | Junyi Ye |
| Abstract: | When a data license expires, deleting stored records does not remove influence encoded in a trained forecaster. Machine unlearning seeks to remove this influence without retraining. We benchmark temporal unlearning with 3, 200 paired references trained on all data and oracles retrained without the requested period. The grid covers five architectures, four rolling folds, five deletable years, and three experimental deletion levels on an S&P 500 volatility panel. The 2020 COVID crisis year produces the largest memorization gap for every architecture. Removing it improves all three deployable models in every fold, with the largest improvement in the 2022 bear market, while the two non-deployable models respond inconsistently. The target for approximate unlearning is the oracle, not low predictive accuracy on the deleted period. In one Transformer cell, an oracle that never trained on 2020 still predicts it at an information coefficient of 0.51, compared with 0.55 for the reference; pushing predictions toward noise reduces test skill. Across twelve deployable architecture-method pairs, only TSMixer with the hinge method remains near the oracle in every fold, closing 74-118% of the reference-to-oracle gap without a measurable loss of test skill. Method rankings vary across architectures and rolling windows. Audit separation rises with prior memorization but can remain small after exact deletion. The window-level loss comparison reaches at most 0.69, and treating stock-level windows as independent inflates the absolute t-statistic by a median factor of 1.9. These results call for an explicit deletion scope, oracle validation for the relevant architecture and window, and power-aware auditing. |
| Date: | 2026–08 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2609.26242 |
| By: | Mehrdad Pournaderi |
| Abstract: | We test whether a multivariate vector X conforms to a specified distribution F, a problem in copula modelling and density forecasting. The Rosenblatt transform reduces it to a test of uniformity, but depends on an arbitrary coordinate ordering that strongly affects power under dependence. We study order randomization: applying the transform under many random orderings and merging the evidence with dependence-robust rules. Reordering conserves the total Mahalanobis signal energy and merely redistributes it, so one ordering is a lucky or unlucky draw. In simulations we observe significant gains in calibrated power over both the expected single random ordering and order-invariant references. Two ingredients are essential: a two-sided base statistic, and a re-estimating parametric bootstrap that restores level under an estimated null and unlocks the gain. The calibrated pooled tests are robust to the departure's shape; no order-invariant reference we compare is: the symmetric-root test collapses on diffuse departures, while the shape-flat chi-squared test trails on concentrated ones. We apply it to a Gaussian foreign-exchange risk model over a decade of daily data on nine currencies, where it detects episodes such as Brexit and COVID. Though we focus on Gaussian nulls, the procedure extends to any null whose conditional distributions can be computed and simulated from. |
| Date: | 2026–09 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2609.37477 |