
Abstract I study how extreme weather events affect macroeconomic expectations to better understand the propagation of major shocks. I identify the immediate and dynamic effects of a Hurricane Katrina-sized shock on business economists’ expectations of GDP growth. While a shock immediately causes a downward revision in expected near-term growth, the total effect, which includes any dynamics due to forecast revision inefficiencies, is more than double the immediate revision. The effects of the shock are also expected to last for up to a year, which contrasts with forecasts from Federal Reserve Board staff and estimates of the actual economic impacts. The relative persistence of the response in expectations is consistent with models of delayed overshooting and highlights the importance of accounting for forecast revision dynamics.
Summary: This paper shows how the structural representation of a vector autoregressive model can support forecast analysis. We offer a unified framework that formalizes how the structural form of the model can help form a narrative for two key statistics in real-time vector autoregressive forecasting: the forecast errors at the outturn of the data, and the consequent revisions of the forecast. We conduct an exercise on the United Kingdom, focusing on the inflation surge that started in 2022. We show that revisions to the inflation forecast were driven not only by contractionary supply-side shocks, but also by a mix of expansionary demand-side shocks, with an important role played by revised past shocks. However, we find that the role of demand shocks was subsequently revised downwards as new data vintages become available.
We adapt double machine learning to macroeconomic time series by combining regularized nuisance estimation with reverse cross-fitting. This deterministic scheme exploits time reversibility to use time-reversed auxiliary blocks and, unlike neighbour-deletion designs, avoids buffer blocks, thereby improving sample usage. We derive conditions for asymptotic validity and show in simulations that the estimator performs well in realistic finite samples across the designs considered, including cases with misspecification, heteroskedasticity, and state dependence. We also show that, in high dimensions, predictive tuning metrics do not minimize bias in the causal score. We therefore propose a calibration rule targeting a Goldilocks zone of tuning parameters delivering stable partialled-out signals and reduced small-sample bias. We extend the method to residualized local projections and apply it to estimate the dynamic effects of a rise in Tier 1 regulatory capital. The results illustrate the usefulness of the approach for macroeconomic time-series inference.
Abstract This paper develops integrated modified ordinary and generalized least squares estimation for systems of cointegrating multivariate polynomial regressions, i. e., systems of equations that include deterministic variables, integrated processes and (cross-)products of non-negative integer powers of these variables as regressors. The stationary errors are allowed to be correlated across equations, over time and with the regressors. The necessity to consider integrated modified generalized least squares estimation arises in the case of estimation under restrictions, which in general implies that ordinary and generalized least squares estimators cease to be identical. We discuss in detail hypothesis testing for the unrestricted and restricted estimators. Furthermore, we develop asymptotically pivotal fixed-b inference, shown to be available only in special cases. A brief illustration to modelling output and emissions as a Cobb-Douglas-Translog cointegrated production system with TFP, capital and labor as inputs indicates the potential usefulness of the developed methods for climate econometrics.
What should a macroeconomically relevant climate index contain? The composition is inherently ambiguous, and aggregation choices affect structural inference. We introduce the Assemblage VAR, which jointly estimates nonnegative aggregation weights and VAR parameters by maximizing the system likelihood-gain criterion, effectively outsourcing aggregation to observed macroeconomic dynamics. Two variants operate in component-space and rank-space, reweighting named subcomponents or emphasizing regions of the cross-sectional distribution. Applied to disaggregated U.S. climate data from the Actuaries Climate Index and NOAA, VARs using the assembled climate measures yield contractionary impulse responses that are substantially larger than those estimated using fixed-weight benchmarks. Weights emphasize high-wind variables and distributional tails over slow-moving components such as sea level.