This study asks whether physical climate-risk transmission across insurance, reinsurance, climate-vulnerable sovereign bond, real-estate, agricultural, water, and adaptation markets is detectable in daily data. A news-based physical-risk index and seven publicly documented market series, observed from January 2020 to June 2026, are analysed with quantile vector autoregression connectedness. Inference uses bias-corrected bootstrap intervals, three placebo nulls, and a Monte Carlo power analysis. The markets are strongly connected (total connectedness 47.0%, 45.1%, and 47.2% at the 5%, 50%, and 95% quantiles), but this connectedness belongs to the equity block and two-fifths of it is a common market factor. Tail amplification is small and disappears once the March 2020 crash is excluded. Reinsurers are net receivers and insurers net transmitters at every quantile. The physical-risk index contributes no more to the markets, or to sovereign bonds in the lower tail, than a placebo index, and rolling estimates show no upward trend. A power analysis shows that the connectedness tests detect only large, linear transmission; a complementary co-exceedance test has 86% power against tail-only transmission above about 2% of receivers’ variance (smaller effects cannot be excluded). Weekly, monthly, and disaster-loss specifications agree. An uncentred innovation covariance mechanically inflates tail connectedness to about 81%.
Whether tourism has recovered from the COVID-19 shock depends on the benchmark used. This study combines interrupted time-series models, structural-break detection, an event-study specification and counterfactual forecasts to quantify the shock and recovery of Greek tourism using 132 monthly observations for 2015–2025. A counterfactual path is forecast from a trend–seasonal model estimated on pre-pandemic data, validated out of sample against seasonal-naïve, exponential-smoothing and SARIMA benchmarks, and compared with SARIMA, damped-trend ETS and growth-scenario alternatives; a sieve bootstrap quantifies the uncertainty of shortfalls and recovery dates. Travel receipts, inbound traveler flows and receipts per traveler are compared. Receipts and arrivals contracted by approximately 83%, a finding that was robust across intervention dates, HAC bandwidths and serially correlated-error specifications, whereas a single constant recovery slope is not supported. Both indicators regained their 2019 levels in April 2023 under all rules. Relative to the counterfactual, receipts met a three-month criterion in February 2024 while arrivals had not by end-2025; receipts cross first with probability 0.64, but the dates depend on the run-length rule and the assumed pre-pandemic growth, and in real terms the post-2022 lead of receipts reflects prices, not higher spending per traveler. Domestic accommodation demand resisted the shock better; foreign demand rebounded faster.
This study investigates whether U.S. macro-financial indicators improve one-step-ahead forecasts of return connectedness among gold, Brent crude oil, and the U.S. Dollar Index beyond autoregressive persistence. We estimate a time-varying Total Connectedness Index (TCI) using a TVP-VAR model with a 10-step generalized forecast-error variance decomposition and historically calibrated initialization. The U.S. Consumer Price Inflation is in line with the scheduled release dates, while predictors without historical data are not considered for the main analysis. Six forecasting models are evaluated using chronological validation and expanding-window estimation over 374 out-of-sample observations (8 May 2024–10 March 2026). The AR(2) benchmark achieves the lowest root mean squared error (RMSE=0.4898, out-of-sample R2=0.9768), compared with 0.4908 for macro-financially augmented linear regression and 0.5313 for XGBoost combining autoregressive and external predictors. The augmented XGBoost model produces 17.65% higher mean squared forecast loss than AR(2) (Holm-adjusted Diebold–Mariano p=2.10×10−5). Rolling-origin SHAP analysis attributes 94.91% of aggregate mean absolute feature attribution to the two lagged TCI predictors, indicating that historical connectedness dominates the attribution structure of the fitted nonlinear forecasting model. Exploratory inflation-regime comparisons do not provide robust evidence of a discrete 3% threshold. Within the evaluated three-market connectedness target, selected macro-financial predictors, one-step forecasting horizon, and 374-observation test period, none of the alternative models improved upon the AR(2) benchmark. These findings highlight the importance of historically consistent target construction, information availability, and persistence-aware forecasting benchmarks. They do not establish that macro-financial information generally lacks predictive value or that nonlinear forecasting models are inherently inferior.
Wind forecasting in complex coastal terrain, such as the Colombian Caribbean, is hindered by air–sea heat fluxes and nonlinear atmospheric dynamics. This pre-registered systematic review and meta-analysis tests whether advanced architectures (artificial intelligence, hybrid, ensemble) outperform classical ones (physical, statistical). Scopus and Web of Science (last query: 22 September 2026) yielded 100 studies with extractable effect sizes; 76 compared a proposed model with an explicit comparator (primary analysis). Hedges’ g (negative values favor the proposed model) was pooled with REML random effects and Knapp–Hartung inference. The pooled effect was g=−1.07 (95% CI [−1.38,−0.75]; 95% prediction interval [−2.77,0.64]; I2=69.5%), robust to the estimator and to influential studies, although 59 comparisons were against a baseline of the same family. The estimates favored advanced over classical architectures, but not significantly (β=−0.30, 95% CI [−0.99,0.38]); only 10 studies compared them directly, and 9 favored the advanced model (g=−0.78, 95% CI [−1.93,0.36]). Only AI (g=−1.08, k=48) and physical (g=−0.93, k=19) models had intervals excluding zero. Egger and Begg tests indicated funnel-plot asymmetry, but the Pustejovsky–Rodgers test, robust to the dependence of the standard error on the effect, did not. Proposed models outperform their baselines; the superiority of any architectural class is not established.
Online conformal prediction methods such as Adaptive Conformal Inference (ACI) and Fully Adaptive Conformal Inference (FACI) adjust prediction intervals under distribution shift, but their calibration is based on a common stream of recent nonconformity scores. We introduce Population-based Adaptive Conformal Ensembles (PACE), a heuristic method that maintains online conformal quantile calibrators with different window sizes, decay rates, and quantile scales. PACE combines the best-calibrated members through fitness-weighted top-K averaging and periodically refreshes the population using clonal selection. For context, Strongly Adaptive Online Conformal Prediction (SAOCP) is a benchmark method that combines online calibration experts operating over different time intervals and provides a formal strongly adaptive regret guarantee. PACE is heuristic and does not provide an analogous regret or coverage guarantee. We evaluate the method on two synthetic datasets and three real-world time series. Against five adaptive conformal baselines, PACE achieves higher empirical coverage during extreme regimes. Compared with SAOCP, it generally obtains higher coverage by producing wider intervals, resulting in less favorable interval scores on most datasets.
