|
on Forecasting |
| By: | Andrew R Garcia (Universidad de Ingeniería y Tecnología); Marco Vega (Banco Central de Reserva del Perú) |
| Abstract: | This paper asks whether structural metadata from an institutional registry carries enough signal to guide variable selection in macroeconomic forecasting. We study this question using two complementary approaches. The first, Metadata ε-Greedy, is a stochastic search policy that uses permutation-invariant embeddings of registry metadata to guide a fixed-budget search over predictor subsets, with forecasting loss as the only feedback signal. The second, Metadata Bayes, performs variable selection entirely within metadata space: it constructs group-level priors from institutional descriptors, updates them via partial correlation with the target, and selects predictors through Thompson sampling, without ever evaluating a forecasting model during selection. Both methods are evaluated on forecasting Peruvian headline CPI under two forecasters, a Vector Autoregression (VAR) and a Random Forest, and benchmarked against random search, greedy forward selection, LASSO, Bayesian Ridge, PCA, and a state-of-the-art Bayesian variable selection method. Metadata Bayes, despite never observing forecasting loss during selection, achieves out-of-sample accuracy competitive with all baselines including the Bayesian benchmark. Metadata ε-Greedy further improves on these results under the VAR during the COVID shock period. Together, the results suggest that registry metadata encodes enough economic structure to serve as a meaningful proxy for predictive relevance, complementing rather than replacing existing forecasting pipelines. |
| Keywords: | Variable selection, Metadata, Economic forecasting, Bayesian methods, Stochastic search |
| Date: | 2026–04 |
| URL: | https://d.repec.org/n?u=RePEc:rbp:wpaper:dt-2026-012 |
| By: | LIU JIENI (Graduate School of Economics, The University of Osaka) |
| Abstract: | This paper proposes a search-then-forecast framework for mid-term (20-trading-day) stock price forecasting and evaluates it on Chinese A-share data. The framework combines a multi-distance voting-based similar-stock search for sample augmentation, a sample-level PCA–ICA feature reconstruction, and a Transformer encoder–decoder whose decoder is initialised at inference with the single-step return at the end of the observation window rather than the conventional zero-padding. Using Luoyang Molybdenum (stock code 603993) from the SSE 180 pool as a single-stock case study, we compare six configurations—TransE, TransED (nhead = 3 and 6), BiLSTM, ARMAGARCH, and a TransED-Embed ablation—across ten observation window lengths, and introduce two baseline-referenced metrics, R2 hist and MASEnaive, to address the limited interpretability of standard R2 and MASE on non-stationary financial series. ARMAGARCH attains the lowest root mean squared error (RMSE) across all tested windows, outperforming the best deep learning model (TransE) by 1.4% to 17.0%; MASEnaive further reveals that most deep learning models fail to surpass a random-walk naive baseline. Observation window length and model architecture exhibit a clear interaction, and a smaller internal Transformer dimension does not hurt performance. Within this single-stock case study, the findings suggest that parsimonious statistical models can match or outperform highly parameterised deep learning architectures for mid-term Chinese A-share forecasting. |
| Keywords: | stock price forecasting; Transformer; ARMAGARCH; similar-stock search; Chinese A-share market. |
| JEL: | C22 C45 C53 G12 G17 |
| Date: | 2026–04 |
| URL: | https://d.repec.org/n?u=RePEc:osk:wpaper:2606 |
| By: | Jennifer Peña; Katherine Jara; Fernando Sierra |
| Abstract: | This paper investigates whether artificial intelligence techniques—encompassing both machine learning and deep learning models—can enhance the accuracy of now-casts for Chile’s monthly economic activity index (IMACEC). The analysis relies on a large and diverse real-time dataset that includes both traditional macroeco-nomic variables and high-frequency monthly administrative data (from electronic tax records). Three main findings emerge. First, nonlinear models—particularly XGBoost—achieve the lowest root mean squared errors, whereas linear regularized approaches such as SVR and LASSO also show competitive performance. This highlights the value of flexible nonlinear methods and regularized linear approaches when dealing with heterogeneous data. Second, features derived from electronic tax records—such as trade credit volumes and sectoral sales by region—consistently rank among the most important predictors across models. Third, the strongest-performing models—XGBoost, SVR, and LASSO—achieve lower errors than tra-ditional econometric benchmarks, which rely solely on standard macroeconomic aggregates and exclude non-traditional datasets. Overall, the findings show that timely administrative data, combined with AI approaches, can significantly improve economic surveillance and decision-making. |
| Date: | 2025–12 |
| URL: | https://d.repec.org/n?u=RePEc:chb:bcchwp:1058 |
| By: | Christian Beer (Oesterreichische Nationalbank, Economic Analysis Division); Robert Ferstl (Off-Site Banking Analysis and Strategy Division); Bernhard Graf |
| Abstract: | This study examines the effectiveness of using webscraped data to predict price developments in the Austrian food retail sector. We calculate monthly nowcasts of price changes based on daily price data collected by the OeNB since mid-2020, using Eurostat methodology for price index calculation, along with further details provided by the national statistics office. We assess the quality of our nowcasts by comparing them with various baseline models and more advanced time series methods also covering machine learning approaches. Our findings indicate that webscraped data are a useful way to obtain more accurate nowcasts with a time advantage, amounting to several weeks, over traditional data sources. In addition, we are the first, to our knowledge, to explore the possibility of using the improved accuracy of the nowcasts as a basis for disaggregated short-term forecasts that extend up to one quarter. While direct forecasts at higher levels of aggregation produce slightly more accurate overall metrics, indirect forecasts derived from disaggregated data provide superior insights into the underlying dynamics of specific sub-components. Our results show that more advanced time series models have trade-offs in terms of computational efficiency while performing very similarly to more traditional methods. These findings have implications for policymakers who aim to develop an effective system for real-time monitoring of inflation dynamics at a very granular level. |
| Keywords: | Webscraping, Inflation forecasting, Time series models |
| JEL: | C22 C81 E31 E37 |
| Date: | 2025–01–16 |
| URL: | https://d.repec.org/n?u=RePEc:onb:oenbwp:262 |
| By: | Schick, Manuel; Opschoor, Anne |
| Abstract: | This paper proposes forecasting joint tail risks for key macroeconomic indicators, GDP growth, inflation, and unemployment, using the US Survey of Professional Forecasters (SPF). By incorporating SPF consensus forecasts into the conditional mean of AR-GARCH-type models, the accuracy of univariate and multivariate predictive densities is significantly improved. Modeling a constant correlation matrix captures strong dependencies, particularly between GDP growth and unemployment. Using US data from 1990 to 2024, we show that the joint modeling framework enables scenario-based analysis in which predictive densities, conditioned on adverse developments in other variables, differ substantially from the baseline marginal distributions. The framework allows for a formal out-of-sample evaluation of joint predictive densities and a transparent assessment of conditional tail risks. |
| Keywords: | Growth-at-Risk; Multivariate Predictive Densities; GARCH; Tail Risk; Macroeconomic Forecasting |
| Date: | 2026–04–24 |
| URL: | https://d.repec.org/n?u=RePEc:awi:wpaper:771 |
| By: | Diego Vivanco Vargas; Camilo Levenier Barría; Lissette Briones Molina |
| Abstract: | High-frequency microdata can significantly enhance the accuracy of nowcasting models for economic activity. This study evaluates the performance of using microdata to nowcast the monthly Chilean activity. We compare models with granular data from electronic invoicing and digital payment systems with conventional univariate and multivariate time series models and leading indicators. For the nowcasts with microdata, we employ SARIMAX specifications and a bottom-up aggregation methodology, complemented with satellite models for specific economic sectors. Our empirical results show a substantial reduction of approximately 34% in root mean square errors (RMSE) for nowcasts of the annual growth of IMACEC (monthly economic activity indicator in Chile) over a 36- month out-of-sample evaluation period. These findings underscore the value of microdata for improving real-time estimates of economic activity, encouraging its integration into nowcasting frameworks. |
| Date: | 2025–12 |
| URL: | https://d.repec.org/n?u=RePEc:chb:bcchwp:1059 |
| By: | Maximilian Göbel (Brain); Philippe Goulet Coulombe (Université du Québec à Montréal); Karin Klieber (Oesterreichische Nationalbank) |
| Abstract: | Machine learning predictions are typically interpreted as the sum of contributions of predictors. Yet, each out-of-sample prediction can also be expressed as a linear combination of in-sample values of the predicted variable, with weights corresponding to pairwise proximity scores between current and past economic events. While this dual route leads nowhere in some contexts (e.g., large cross-sectional datasets), it provides sparser interpretations in settings with many regressors and little training data—like macroeconomic forecasting. In this case, the sequence of contributions can be visualized as a time series, allowing analysts to explain predictions as quantifiable combinations of historical analogies. Moreover, the weights can be viewed as those of a data portfolio, inspiring new diagnostic measures such as forecast concentration, short position, and turnover. We show how weights can be retrieved seamlessly for (kernel) ridge regression, random forest, boosted trees, and neural networks. Then, we apply these tools to analyze postpandemic forecasts of inflation, GDP growth, and recession probabilities. In all cases, the approach opens the black box from a new angle and demonstrates how machine learning models leverage history partly repeating itself. |
| Date: | 2025–03–27 |
| URL: | https://d.repec.org/n?u=RePEc:onb:oenbwp:265 |
| By: | Winkelried, Diego (Universidad del Pacífico); Jason Cruz (Universidad del Pacífico); Javier Torres (Universidad del Pacífico) |
| Abstract: | This paper analyzes the revision process of monthly GDP growth in Peru using a newly constructed real-time dataset. A unified empirical framework is developed to test rationality by jointly examining serial and vintage correlations in revisions. The results reveal systematic and predictable revision patterns, primarily driven by benchmarking procedures followed by the statistical agency. Motivated by these findings, a simple forecast-adjustment model is proposed that improves early assessments of economic activity by anticipating subsequent revisions. The evidence illustrates how revision dynamics in an emerging economy can be exploited to enhance nowcasting performance. |
| Keywords: | Data revisions, rationality tests, nowcasting, real-time data, emerging markets. |
| JEL: | C22 C53 C82 E01 O11 O54 |
| Date: | 2026–04 |
| URL: | https://d.repec.org/n?u=RePEc:rbp:wpaper:dt-2026-003 |
| By: | Elías Albagli; Guillermo Carlomagno; Javier Ledezma; María Teresa Reszczynski |
| Abstract: | We propose a structural framework to uncover the key forces shaping global asset prices and financial conditions. Our approach identifies seven distinct shocks: four U.S.-centric (growth, monetary policy, common risk premium, and a novel dollar-hedging risk), alongside a global hedging risk premium, a China-growth shock and an emerging market-specific risk premium shock. Using daily financial data from 2010–2025, we estimate a Structural VAR to trace how these shocks propagate across advanced and emerging economies. Our contributions are threefold. First, we introduce a real-time monitoring tool that provides structural interpretation and scenario analysis, equipping policymakers with a unified lens to assess asset price dynamics. Second, we improve shock identification through three innovations: (i) incorporating the dollar-hedging risk shock to explain anomalies observed since 2025, (ii) improving U.S. shock identification by leveraging non-U.S. data, and (iii) highlighting the pivotal role of Chinagrowth shocks in shaping emerging-market conditions. Finally, we develop a novel Financial Conditions Index (FCI) grounded in structural shocks, enabling country-specific assessments and enhancing interpretability. Unlike traditional FCIs, our index directly links financial conditions to their economic drivers, improving realtime monitoring and outperforming existing alternatives in nowcasting economic activity. |
| Date: | 2026–04 |
| URL: | https://d.repec.org/n?u=RePEc:chb:bcchwp:1080 |
| By: | Bowden, Roger; Zhu, Jennifer; Cho, Jin Seo |
| Abstract: | Directional calls are often more successful than precise value prediction, particularly at certain times, when underlying fundamentals suggest a breakout from the stable range. We adapt the categorical directional framework implicit in binomial or trinominal step processes to establish nonhomogeneous multinomial directional probabilities over coarser time intervales and show how such frameworks can be used for forecasting the hedging, including dynamic persistence. Problems of signal compression and outcome definition can be addressed using methods analogous to neuronal nets and fuzzy membership functions. the methods are applied to derive forecasting and conditional hedge procedures for foreign exchange exposures. |
| Keywords: | Conditional value at risk, Foreign exchange forecasting, Fuzzy regimes, Hidden Markov models, Non-homogenous multinomial process, |
| Date: | 2026 |
| URL: | https://d.repec.org/n?u=RePEc:vuw:vuwecf:33487 |