Research

Summary: When agents' information is imperfect and dispersed, existing measures of macroeconomic uncertainty based on the forecast error variance have two distinct drivers: the variance of the economic shock and the variance of the information dispersion. The former driver increases uncertainty and reduces agents' disagreement (agreed uncertainty). The latter increases both uncertainty and disagreement (disagreed uncertainty). We use these implications to identify empirically the effects of agreed and disagreed uncertainty shocks, based on a novel measure of consumer disagreement derived from survey expectations. Disagreed uncertainty has no discernible economic effects and is benign for economic activity, but agreed uncertainty exerts significant depressing effects on a broad spectrum of macroeconomic indicators.

Summary: Composite quantile regression (CQR) is a robust and efficient estimator under heavy-tailed and contaminated errors. Existing Bayesian extensions rely on working likelihoods that require latent-variable augmentation and can deliver poorly calibrated credible intervals. We develop generalized Bayesian CQR, which exponentiates the composite quantile loss directly, targeting the same objective as frequentist CQR. Because generalized Bayes replaces point optimization with posterior averaging over the loss surface, it is especially relevant under heavy-tailed errors where the composite quantile loss flattens near its minimum. In generalized Bayes posterior dispersion depends on a learning rate that we calibrate by matching marginal variances to their frequentist sandwich counterparts. The resulting credible intervals achieve near-nominal coverage in cross-sectional settings and substantially reduce the undercoverage of i.i.d. intervals under serial dependence, with a residual shortfall under high persistence that mirrors the finite-sample bias of frequentist HAC inference. The calibration has a closed-form solution under flat priors and extends to normal and spike-and-slab LASSO priors for shrinkage and variable selection. Sampling uses standard Metropolis-Hastings with no latent variables, achieving roughly 100-fold computational gains over likelihood-based Bayesian CQR at a common quantile grid. Monte Carlo experiments show competitive or improved point estimation relative to frequentist CQR, reliable coverage, and robust variable selection across Gaussian, heavy-tailed, and contaminated error distributions. An equity premium forecasting application demonstrates that the efficiency and robustness gains translate into economically meaningful improvements in out-of-sample portfolio performance.

Summary: We develop a Quantile Bayesian Vector Autoregression (QBVAR) to forecast real oil prices across different quantiles of the conditional distribution. The model allows predictor effects to vary across quantiles, capturing asymmetries that standard mean-focused approaches miss. Using monthly data from 1975 to 2025, we document three findings. First, the QBVAR improves median forecasts by 2-5% relative to Bayesian VARs, demonstrating that quantile-specific dynamics matter even for point prediction. Second, uncertainty and financial condition variables strongly predict downside risk, with left-tail forecast improvements of 10-25% that intensify during crisis episodes. Third, right-tail forecasting remains difficult; stochastic volatility models dominate for upside risk, though forecast combinations that include the QBVAR recover these losses. The results show that modeling the conditional distribution yields substantial gains for tail risk assessment, particularly during major oil market disruptions.

Summary: We develop a functional-coefficient time-varying parameter vector autoregression (TVP-VAR) for macroeconomic forecasting, in which coefficients evolve as deterministic functions of observable exogenous variables, permitting straightforward linear estimation. By allowing economic indicators to signal structural change directly, the model disciplines parameter variation more effectively than likelihood-based alternatives, while avoiding the dimensionality, prior sensitivity, and computational costs of standard TVP-VARs. Among the specifications considered, including polynomials, interactions, and machine-learning methods, a parsimonious linear form with persistent coefficients performs best. Simulations and applications to U.S. and euro-area macroeconomic data show that the proposed adaptively-varying parameter VAR delivers consistently superior out-of-sample forecasts, especially during volatile periods.

Summary: I introduce a high-dimensional Bayesian vector autoregressive (BVAR) framework designed to estimate the effects of conventional monetary policy shocks. The model captures structural shocks as latent factors, enabling computationally efficient estimation in high-dimensional settings through a straightforward Gibbs sampler. By incorporating time variation in the effects of monetary policy while maintaining tractability, the methodology offers a flexible and scalable approach to empirical macroeconomic analysis using BVARs, well-suited to handle data irregularities observed in recent times. Applied to the U.S. economy, I identify monetary shocks using a combination of high-frequency surprises and sign restrictions, yielding results that are robust across a wide range of specification choices. The findings indicate that the Federal Reserve's influence on disaggregated consumer prices fluctuated significantly during the 2022-24 high-inflation period, shedding new light on the evolving dynamics of monetary policy transmission. 

Summary: From the perspective of flexible inflation targeting using a simple targeting rule, this paper introduces the Monetary Policy Deviation Error (MPDE) as a novel metric for assessing central bank performance and deliberations. The MPDE captures potentially time-varying shifts in the trade-off between stabilizing inflation and supporting real economic activity. Specifically, it quantifies the gap between the intended trade-off envisioned by policymakers and the trade-off realized through actual monetary policy outcomes. Under an optimal and unbiased monetary policy strategy, the MPDE should average to zero. Nonzero deviations indicate misalignment between the central bank's stated objectives and the trade-offs actually achieved, suggesting that an alternative interest rate path would have better aligned outcomes with intentions. Applying the MPDE to evaluate the monetary policy strategies of Norges Bank and the Reserve Bank of New Zealand, we find posterior evidence supporting optimal policy alignment in the case of New Zealand.