|
on Forecasting |
| By: | Frédérqiue Bec; Heino Bohn Nielsen (CY Cergy Paris Université, THEMA) |
| Abstract: | Forecast error variance decompositions (FEVDs) are widely used to assess the contribution of structural shocks in vector autoregressions. However, many variables of interest are nonlinear functions of underlying variables, rendering the standard linear FEVD incomplete. We develop a framework for variance decomposition of nonlinear forecast targets in terms of the Shapley value decomposition and compare it with more conventional approaches based on Taylor expansions. We illustrate that nonlinear interaction effects can account for components of forecast uncertainty that are not fully captured by Taylor approximations. As a result, approximation-based FEVD may substantially distort the picture of forecast uncertainty and the attribution of variance across shocks. |
| Keywords: | Vector Autoregression; Nonlinear Forecast Error Variance Decomposition; Shapley Shares; Generalized Shapley Shares; Interaction Terms. |
| JEL: | C32 C13 E44 |
| Date: | 2026 |
| URL: | https://d.repec.org/n?u=RePEc:ema:worpap:2026-09 |
| By: | Arkadiusz Lipiecki; Rafa{\l} Weron |
| Abstract: | Foundation models promise accurate forecasts with little or no task-specific training, but whether they can replace models designed specifically for electricity price forecasting remains unclear. We compare nine variants from five foundation model families, evaluated in zero-shot mode, with two state-of-the-art electricity price forecasting benchmarks in Germany, Poland, and Spain over 2021-2025. Their performance is assessed in terms of point and probabilistic forecasting accuracy, as well as economic value in battery energy storage arbitrage. Only the TabPFN models consistently and significantly outperform the benchmarks across all three markets and all statistical measures. However, this statistical dominance does not translate directly into economic dominance: TabPFN performs best under unlimited bids and riskier quantile-based strategies, whereas the Distributional Deep Neural Network benchmark is more profitable when risk tolerance is lower. Thus, foundation models cannot universally replace market-specific models, and their value depends on both model architecture and the decision problem. |
| Date: | 2026–08 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2609.00089 |
| By: | Haiqiang Chen; Li Chen; Yunlong Chen; Difang Huang; Bo Zhang |
| Abstract: | We examine whether post-origin training information inflates the measured accuracy and economic value of financial forecasts. We evaluate five sets of financial time-series foundation models, each comprising independently trained annual vintages under U.S., global, and factor-augmented training environments, across 14 equity markets and four forecast horizons. Rolling comparisons vary the annual vintage for a fixed forecast; fixed-vintage comparisons hold the vintage fixed as target windows move across its training cutoff. Each alternative forecast is paired with an origin-aligned point-in-time (PIT) benchmark using identical numerical histories and inference protocols. In the U.S.-trained reference environment, post-origin vintages materially revise informative PIT forecasts but generally reduce accuracy in both designs. Pooled rolling comparisons yield higher mean squared forecast errors in 18 of 20 U.S. model-set-horizon combinations. The origin-crossing update also performs worse on average than an equally long pre-origin update. Under a common constrained allocation rule using one-month forecasts, median exposed-minus-PIT differences in annualized certainty-equivalent returns are -1.77 percentage points in the United States and -2.14 points internationally. Global and factor-augmented training produce more mixed predictive effects. An exact squared-error decomposition shows that revisions improve accuracy when their error-correcting benefit exceeds their mean squared magnitude; under U.S. training, alignment with PIT errors generally falls short of this requirement. Temporal exposure therefore establishes an information-set violation, not sufficient evidence of inflated predictive accuracy or investor value. |
| Date: | 2026–09 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2609.20554 |
| By: | Vegard H. Larsen; Leif Anders Thorsrud |
| Abstract: | Central banks publish projection paths, but these need not be optimal summaries of the information used to form them. We reframe conditional macroeconomic forecasting as an input problem: a fixed, pre-trained multivariate foundation time-series model reads the institution's historical predictions and announced future paths alongside target histories, rather than imposing the path as a model-consistent restriction inside a locally estimated system. The mapping nests the Mincer-Zarnowitz and Granger-Ramanathan regressions as its parametric-linear special case. Reading just Norges Bank's published path triple and target histories, the map cuts mean squared error against the Bank on inflation across horizons, and the nested combination regression puts the conditional weight on the map, not the path. A hard-conditioned VAR matched on the same future paths, but blind to the prediction record, is consistently outperformed on the rate and inflation, and the result survives the asymmetric-loss specifications that best rationalise the path. The VAR contrast replicates on Sweden and New Zealand; the institutional comparison only on New Zealand, with parity at best on Sweden. An institutional-learning regression rationalises the split: the Riksbank absorbs its recent misses more aggressively than Norges Bank and the RBNZ, leaving less residual signal to extract. |
| Keywords: | foundation models, conditional forecasting, central bank forecasts, information efficiency, Chronos-2 |
| JEL: | C53 E37 E47 |
| Date: | 2026 |
| URL: | https://d.repec.org/n?u=RePEc:ces:ceswps:_12972 |
| By: | Parsa Yousefnezhad; Gholamreza Mansourfar; Mohammadreza Feizi Derakhshi |
| Abstract: | Bitcoin's future fluctuations are a substantial concern for investments and risk management. Investors and financial institutions require accurate forecasts of these price movements to hedge and optimize portfolios. This study aims to answer the question of whether Bitcoin's price will rise beyond 5% within the next 7 days by utilizing on-chain, market, and sentiment data from February 2018 to December 2025. The proposed model consists of a multi-scale temporal convolutional network with InceptionTCN blocks, CNN channel attention, adaptive average pooling, and a pairwise ranking loss. Dilated convolutions with bottleneck and fusion layers are employed to efficiently capture features over horizons from 1 to 4 days. Given the class imbalance in the dataset, AUC is used instead of accuracy and other classification metrics to reflect the model's performance better. Subsequently, a profit-optimized decision threshold is also applied to align model selection with financial objectives. The proposed model is compared with 5 other baselines: ImprovedTCN_GRU, LSTM, TCN, XGBoost, and Random Forest. Results indicate that the proposed model achieved an AUC of 0.6316 and a profit of 1.703, outperforming all baseline models. Using a novel deep learning model would assist investors in making better financial decisions. |
| Date: | 2026–07 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2608.26174 |
| By: | Tomasz Weron; Katarzyna Maciejowska |
| Abstract: | This article examines various methods of constructingmultidimensional probabilistic forecasts of electricity prices. Building on the Multiple Split (MS) method, it incorporates forecast averaging across estimation windows of different lengths and compares its performance with that of other, well-established methods. The research demonstrates that the ensemble representation of the price distribution is particularly useful in battery energy storage system (BESS) management. It enables the direct construction of probabilistic forecasts of daily profits. These forecasts can be used to determine optimal charging and discharging hours, as well as to support risk management decisions. The methods are evaluated using data from the German and Spanish day-ahead electricity markets from 2021-2024. The results indicate that theMS method with averaging (MS-ave) generally outperforms the other considered approaches in terms of Prediction Interval Coverage Probability (PICP), the Continuous Ranked Probability Score (CRPS) and the Energy Score (ES). Moreover, it is superior in supporting BESS trading strategies, particularly in case of non-zero operational costs. |
| Date: | 2026–06 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2608.26122 |
| By: | Ayla Jungbluth (Ruhr-University Bochum); Johannes Lederer (University of Hamburg); Simon Trimborn (University of Amsterdam) |
| Abstract: | Modeling the joint distribution of extreme values in high-dimensional financial time series is challenging because extremes are sparse and locally extreme observations are not necessarily extreme relative to their full marginal distribution. To address this, we introduce a time-dependent network Hüsler-Reiss model in which market-informed adjacency matrices determine how strongly observations contribute to the estimation. We propose binary and weighted specifications, including the Joint Extremes Adjacency Matrix (JEAM) which combines information about individual extremeness with historical patterns of joint extreme movements. In the forecasting evaluation part, covering one-minute stock returns from three sectors of the S&P 100, JEAM achieves the best out-of-sample log scores for both tail directions; improving scores by 12.5-13.6% in the lower tail and 11.4-14.9% in the upper tail. The results show that incorporating market-informed network structures in the estimation, improves forecast evaluation of extremes across time series. |
| JEL: | C53 C58 G17 |
| Date: | 2026–09–13 |
| URL: | https://d.repec.org/n?u=RePEc:tin:wpaper:20260070 |
| By: | Sean Wan; Dongping Liu; Luyao Zhang |
| Abstract: | We introduce StableEval Arena, a cost-aware benchmark framework for evaluating agentic AI systems on stablecoin peg-risk prediction. StableEval Arena evaluates LLM-backed agentic systems on diagnosing peg stress and forecasting deviations from the one-dollar peg over a hidden seven-day horizon, using leakage-safe historical replay with exchange price-volume data and market-context features. We report two complementary experiment blocks: a 120-case stress-enriched validation block and a 507-case natural-distribution full-arena evaluation block. Across six LLM-backed agent configurations and baselines, StableEval Arena measures prediction quality, calibrated-label behavior, structured-output reliability, latency, token consumption, and estimated inference cost. Rather than ranking agents by accuracy alone, the framework treats trustworthiness as a joint property of forecast quality, operational reliability, and computational cost. The results show a gap between protocol-following reliability and financial-risk reliability: agents reliably produce valid structured outputs at modest measured cost, but still miss most rare severe-stress and sustained-depeg cases. To support auditing and replication, we release the benchmark dataset on Hugging Face and the source code on GitHub. |
| Date: | 2026–08 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2609.18949 |
| By: | Anastasios Megaritis (HUBS - Hull University Business School - University of Hull [United Kingdom]); Dimitrios Bakas (Nottingham Trent University); Theodora Bermpei (LEM - Lille économie management - UMR 9221 - UA - Université d'Artois - UCL - Université catholique de Lille - ULCO - Université du Littoral Côte d'Opale - Université de Lille - CNRS - Centre National de la Recherche Scientifique); Athanasios Triantafyllou (Audencia Business School) |
| Abstract: | In this paper, we examine the macroeconomic predictive power of the volatility of the US Treasury yield curve slope (term spread volatility). Our forecasting exercise shows that US term spread volatility has significant predictive power for US industrial production and employment growth. The predictive power of term spread volatility is stronger at medium‐ and long‐term forecasting horizons and remains robust when well‐established predictors of economic activity, such as the term spread and stock market returns, are included. Our results also show that term spread volatility has statistically and economically distinct predictive power compared to other measures of economic uncertainty. Moreover, the predictive power of term spread volatility increases significantly after the 2008 Great Recession, indicating that the relationship between uncertainty about macroeconomic expectations and macroeconomic performance has strengthened in the post–Great Recession period. Finally, our out‐of‐sample forecasting results show that term spread volatility outperforms the term spread in forecasting economic activity over the longer term. |
| Keywords: | volatility, term spread, forecasting, economic activity |
| Date: | 2026–08 |
| URL: | https://d.repec.org/n?u=RePEc:hal:journl:hal-05740995 |
| By: | Nneka Umeorah; Tolulope Fadina |
| Abstract: | This paper develops a unified framework for assessing systemic risk and identifying contagion channels in the global banking system using a Temporal Heterogeneous Multiplex Graph Neural Network. We construct a harmonised quarterly panel combining bank fundamentals, CDS spreads, and macroeconomic indicators, and represent these data as dynamic multiplex networks linking banks through financial similarity and liquidity co-movement, augmented with country-level macroeconomic relationships. The model integrates graph convolutional layers with recurrent GRU dynamics and incorporates a learnable fusion gate to capture time-varying reliance on alternative contagion channels. Empirical results show that the framework outperforms conventional econometric, machine learning, and graph-based benchmarks for short-term changes in CDS spreads. Beyond forecasting, we provide an interpretable framework to quantify bank-level systemic importance via stress testing, assess country-level spillovers under macroeconomic shocks, and uncover transmission pathways through edge perturbation analysis. Robustness tests confirm the stability of both predictive accuracy and systemic risk rankings. |
| Date: | 2026–08 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2608.27295 |
| By: | Sayan Dhan; Selvaraju Natarajan |
| Abstract: | Formulaic alpha discovery is a pool-dependent symbolic search problem in which informative feedback is observed primarily when a complete expression is evaluated. This delayed feedback creates two coupled difficulties: the retained alpha pool does not preserve the full history of realized evaluation feedback, and the value of an intermediate construction action is uncertain because its consequence depends on the formula eventually completed. We introduce AlphaRJM, which addresses these difficulties through Reward-Jump Memory, an event-driven latent state that remains fixed during token construction and updates only at terminal evaluation events using the realized pool reward and evaluation outcome, and an action-conditioned SDE return critic that represents future discounted discovery returns with stochastic particles. The particles guide action selection through their mean and uncertainty and are learned using a distributional Bellman objective combining energy-distance matching, mean calibration, and jump regularization. Empirically, AlphaRJM delivers strong and stable gains across multiple equity universes, forecasting horizons, and random seeds, while ablations confirm the complementary roles of persistent evaluation history, stochastic return modeling, and distributional supervision. |
| Date: | 2026–09 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2609.08581 |
| By: | Chuanzhen Wang; Alice Zhang; Wei Chen; Michael Brown |
| Abstract: | Accurate forecasting of realized volatility ($RV$) is crucial for risk management and derivatives pricing. Although the implied volatility ($IV$) surface offers rich informational content, prevailing methods that treat it as a static image fail to capture its inherent dynamics. To overcome this limitation, we propose the Finance-Aware Graph Spatio-Temporal Network (FA-GSTN), a novel architecture that reframes $RV$ forecasting as modeling the evolution of a structured financial object. FA-GSTN builds a spatio-temporal graph sequence from the $IV$ surface, where nodes correspond to grid points and edges encode adaptive spatial (intra-day) and explicit temporal (inter-day) dependencies. The model incorporates domain knowledge through finance-aware node features (e.g., option Greeks) and tackles high-frequency noise via a multi-scale temporal smoothing gate coupled with an adaptive robust loss function. Comprehensive evaluations on a large-scale equity options dataset show that FA-GSTN sets a new state of the art, delivering superior predictive accuracy ($R^2$ up to 0.473). It also demonstrates remarkable data efficiency, substantially outperforming strong Vision Transformer baselines when trained on only one year of data ($R^2$: 0.372 vs. 0.315). Furthermore, the model exhibits enhanced robustness during periods of market stress, such as 2020--2021. Ablation studies confirm the vital roles of the spatio-temporal graph structure, finance-aware components, and integrated noise-handling modules. Our work underscores the substantial benefits of explicitly modeling temporal dynamics and infusing financial inductive biases for accurate and robust volatility forecasting. |
| Date: | 2026–06 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2608.26127 |
| By: | Luke Hartigan |
| Abstract: | I develop a statistical measure of Australia's common output cycle by combining frequency-domain filtering of industry-level output to isolate its cyclical component with a dynamic factor model featuring stochastic cycle dynamics. The common cycle has an estimated period of around six years and reveals substantial heterogeneity in the timing and strength with which industries participate in the common cycle. Manufacturing and wholesale trade are closely synchronised, while agriculture and mining are mostly driven by their own idiosyncratic cycles. I evaluate the common output cycle's ability to forecast inflation and find that its predictive performance is statistically indistinguishable from that of an AR(1) benchmark over the forecast horizons considered. I also find evidence of potential endpoint issues when estimating the common output cycle in real time. Overall, these findings suggest that the common output cycle is best viewed as a descriptive measure of the historical features of the Australian output cycle that are shared across industries. |
| Keywords: | business cycle measurement, Butterworth filters, dynamic factor model, inflation forecasting, Kalman filter, signal extraction, stochastic cycle |
| JEL: | C32 C38 C53 E32 E37 |
| Date: | 2026–09 |
| URL: | https://d.repec.org/n?u=RePEc:een:camaaa:2026-78 |
| By: | Joshua Le Grice |
| Abstract: | Algorithmic trading now represents a market exceeding $20 billion, where even marginal gains in signal robustness can translate into economically significant returns. Existing evaluations of equity prediction models do not explicitly target regime robustness during hyperparameter selection. Five model classes are trained on daily observations from approximately 300 large-cap US equities over eleven years, with Bayesian optimisation configured to target trading performance across three statistically different market regimes. Regime-robust hyperparameter selection is associated with out-of-sample generalisation, as signal precision remains above the random baseline across all four quarters of the test period, and portfolio performance slowly degrades under simulated input noise before collapsing beyond a defined threshold. No individual tabular deep learning architecture outperforms gradient-boosted trees, but combining XGBoost and TabNet using rank aggregation produces a Hybrid ensemble with an annualised return of 51.26%, a Sharpe ratio of 2.44, and a statistically significant CAPM alpha of 0.423 (p = 0.011). A near-zero beta indicates this outperformance is driven by stock selection, not market exposure. Alternative data plays a secondary role once technical and fundamental features are accounted for, as well as contributing more strongly on the short side than the long, and varies by model class. An interactive application makes these results explorable in real time, with live data integration the remaining step toward practical deployment. |
| Date: | 2026–08 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2608.27076 |
| By: | Irma Alonso-Alvarez (Banco de España); Iván Kataryniuk (Banco de España); David López-Salido (Banco de España); Javier J. Pérez (Banco de España) |
| Abstract: | This paper develops a tractable framework that maps geopolitical risk in the Middle East to oil prices and macroeconomic conditions through three linked blocks. Geopolitical uncertainty is modelled as a probability distribution over a small set of mutually exclusive scenarios, whose weights are updated with prediction-market prices, structured expert assessments and high-frequency indicators via Bayesian methods. Each scenario is then translated into a net oil supply shortfall after inventory and reserve buffers, which in turn feeds a reduced-form but economically interpretable Brent price equation grounded in recent oil market research. The same architecture can be combined with standard statistical forecasts to check the internal consistency of scenario mixtures and, in reverse, to infer the scenario configuration implicit in observed oil prices. The framework is designed as an operational tool for central banks and other policy institutions to structure geopolitical scenario analysis, quantify resulting oil price and inflation risks and update these assessments in real time. |
| Keywords: | geopolitical risk, oil supply disruptions, energy security, scenario analysis, Bayesian updating, oil prices, inflation, macroeconomic projections |
| JEL: | C11 C53 E31 E37 F51 Q31 Q43 |
| Date: | 2026–09 |
| URL: | https://d.repec.org/n?u=RePEc:bde:opaper:2616e |
| By: | Manuel Naviglio; Fabrizio Lillo |
| Abstract: | Understanding the joint dynamics of prices and trades is central to market microstructure, where returns and order flow interact through nonlinear and state-dependent mechanisms. Linear models are interpretable but may miss these effects, while deep neural networks improve forecasting at the cost of transparency. We use neural networks as tools for structural discovery rather than only for prediction. A deep feed-forward network is trained on high-frequency returns and signed volumes for large- and small-tick stocks and compared with a linear VAR benchmark. The neural network improves predictive performance, especially for returns, revealing nonlinear dependencies beyond the linear specification. Using Shapley-based explainability, we show that the dominant contributions are concentrated at the most recent lags. Model-implied responses are consistent with conditional averages reconstructed from the data. Unlike empirical averages, however, the neural-network decomposition isolates individual regressor contributions to the aggregate dependence. Lagged signed volume generates sign-preserving and saturating effects, consistent with nonlinear price impact and order-flow persistence. Lagged returns act as state variables: when the previous trade does not move the price, the model predicts continuation in the direction of past order flow, whereas non-zero returns generate attenuation or reversal. Building on these findings, we introduce a parsimonious SHAP-inspired nonlinear parametric model. It reproduces the main return-volume dependencies, outperforms the linear VAR benchmark, and achieves performance comparable to the neural network. A multi-lag extension captures residual longer-memory effects while preserving interpretability. Overall, explainability offers a route from black-box prediction to economically meaningful parametric models of price and trade dynamics. |
| Date: | 2026–09 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2609.06085 |
| By: | Baichuan Li; Mengxiao Wang |
| Abstract: | We re-estimate canonical option-implied predictability evidence using a unified 2015--2026 panel of 12.36 million U.S. equity firm-day observations across 10, 026 underlyings. We split the sample into three regimes: late-post-crisis low volatility (2015--2019), high-volatility transition (2020--2022), and AI/mega-cap concentration (2023--2026). The Xing et al. (2010) smirk--return relationship weakens steadily: the next-month univariate smirk coefficient falls from $-0.023$ $(t=-5.5)$ in 2015--2019 to an insignificant $-0.006$ $(t=-1.5)$ in 2023--2026, and turns positive at the three-month horizon $(+0.016, \ t=+2.1)$. In joint specifications with all six canonical signals, the smirk is insignificant throughout and again changes sign in the latest regime. By contrast, the Cremers--Weinbaum IV spread and a Bakshi et al. (2003)-style risk-neutral skewness measure remain significant across regimes and specifications. A boosted-tree benchmark using IV-surface, trading-activity, Greeks, and liquidity features outperforms linear models for next-month return prediction only in the AI/mega-cap regime, with $R^2_{\mathrm{OOS}}=+1.29%$ versus $+0.07%$. Permutation importance identifies different leading predictors by regime, and no canonical hand-engineered signal enters the top five. For firm-level five-day crash classification, AUC is highest in calm markets $(0.706, \ \text{XGBoost})$ and lowest during the high-volatility transition $(0.561)$. Overall, option-implied predictability is real but regime-dependent, and the post-2023 AI/mega-cap period differs sharply from the pre-COVID setting in which the canonical results were established. |
| Date: | 2026–06 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2608.26115 |
| By: | Alex Chen |
| Abstract: | We study statistical predictability in daily U.S. equity prices using only information available at the market open. Using SPY from February 1, 1993 through March 15, 2024, we benchmark XGBoost against Random Forest, LightGBM, Logistic Regression, and naive baselines under expanding-window walk-forward validation. After observing the current day's opening price and two lagged target-specific prices, the task is to predict whether the same day's close will be above or below the previous day's close. On the last 800 trading days, Logistic Regression attains the highest close-direction accuracy (71.09%), Random Forest reaches 61.20%, and XGBoost reaches 58.45% with a 95% bootstrap confidence interval of [54.94%, 62.08]. For XGBoost, close-direction accuracy rises to 72.7% when the predicted move exceeds 1%, but the usable sample falls to 154 observations. We also report Diebold-Mariano and McNemar tests, regime-specific results, SHAP feature importance, and an auxiliary 541-equity screen. The evidence supports a narrow statistical-finance conclusion: simple daily equity features contain detectable same-day directional information, but the result should be interpreted with careful sample-size accounting and limited economic claims. |
| Date: | 2026–05 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2608.26106 |
| By: | Michael Stanley Smith; Lin Deng |
| Abstract: | Multivariate longitudinal data may exhibit non-Gaussian margins, nonlinear dynamics, and response vectors with composition that varies across waves. To account for these features, we introduce a vector drawable vine (VD-vine) copula that extends conventional drawable vine copulas from scalar to vector-valued nodes. Here, the response vector at each wave forms a multivariate marginal, and serial dependence is captured through a sequence of linking vector copulas. We establish that the VD-vine is itself a vector copula and reduces to a conventional drawable vine for scalar nodes. Recursive forward and backward conditional transports are derived that enable efficient likelihood evaluation and predictive simulation, with parsimonious reductions under finite-order Markov and stationary restrictions. Unconstrained parameterizations for Gaussian and FGM linking vector copulas, flexible multivariate marginals, and Bayesian variational inference provide a practical implementation. Simulations show improved predictive accuracy when the marginals are asymmetric and serial dependence is multivariate, with little loss under a correctly specified Gaussian panel vector autoregression. In an eight-wave Australian panel of 1, 093 individuals with varying response vectors, the full VD-vine delivers the best cross-validated distributional forecasts among the models considered, establishing the benefit of capturing asymmetry and nonlinear dependence. |
| Date: | 2026–09 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2609.19547 |
| By: | Hesandi Mallawarachchi; Senilka Madurapperumage; Nadil Kulathunge; Thilokya Angeesa; Nethsith Gunaweera; Sandeepa Weerasekara; Patalee Narasinghe; Nisansa de Silva; Sandareka Wickramanayake |
| Abstract: | The Colombo Tea Auction (CTA) plays a vital role in determining global tea prices, yet the relationship between local weather conditions and price behavior across different tea catalogues has not been thoroughly explored. In this study, we develop a novel, structured dataset by extracting information from 105 weekly broker reports spanning late 2023 to 2026, and combined with region-specific weather data. Our analysis focuses on four main tea catalogues of Sri Lankan tea: High Grown, Low Grown, Off-Grade, and Dust. To better understand the factors influencing tea prices, we apply Granger causality analysis alongside tree-based machine learning models: Random Forest, XGBoost, LightGBM, and Gradient Boosting. Our results show that while market dynamics are primary drivers, weather conditions also have significant effects. Notably, Low Grown tea shows strong sensitivity to precipitation and sunshine duration (p |
| Date: | 2026–06 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2608.24894 |