MACROCAST is the first leakage-free time series foundation model for real-time macroeconomic forecasting, trained exclusively on synthetic series and vintage data, outperforming AR(1), Chronos-2, BVAR, and DFM benchmarks on FRED-MD.
Title resolution pending
10 Pith papers cite this work. Polarity classification is still indexing.
10
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
verdicts
UNVERDICTED 10roles
background 1polarities
background 1representative citing papers
Proposes variance deltas, an interactive tree-based visualization system that identifies subsets of unobserved quantities explaining posterior uncertainty, with demonstrations on causal inference and polling data.
Develops consistent procedures and an efficient alternating least squares algorithm for determining the number of dynamic factors and filter length in dynamic factor models, applied to US macroeconomic time series.
A new test statistic and bootstrap for independence testing of high-dimensional nonstationary time series that avoids whitening by removing temporal dependence bias under the null.
A neural doubly robust proxy causal learning framework using mean embeddings for treatment bridges provides consistent estimators for causal dose-response functions under unobserved confounding for continuous and structured treatments.
Adaptive specification search in financial machine learning produces statistically significant backtests even when no predictability exists, and a new audit using synthetic null environments plus an absolute magnitude gap can detect and quantify such spurious results.
Proposes adaptive and alternative algorithms to improve the computational efficiency of simulation smoothing for large mixed-frequency VARs in nowcasting applications.
Causal inference framework applied to natural experiments measures coupon timing effects on engagement, shown on company onboarding data and a public retention dataset.
SPICE is a scalable Bayesian MCMC engine for explanatory IRT calibration on sparsely linked persons and items in large assessment banks.
Modifies Gibbs sampler for GP state-space models, introduces CFA measurement structure, and validates software via simulation-based calibration to enable reliable learning of nonlinear latent dynamics.
citing papers explorer
No citing papers match the current filters.