Working-Papers 

Serial-Dependence and Persistence Robust Inference in Predictive Regressions, 2026,  [Supplementary] [Econometric Theory, forthcoming]

Abstract: This paper introduces a new method for testing the statistical significance of estimated parameters in predictive regressions. The approach features a new family of test statistics that are robust to the degree of persistence of the predictors. Importantly, the method accounts for serial correlation and conditional heteroskedasticity without requiring any corrections or adjustments. This is achieved through a mechanism embedded within the test statistics that effectively decouples serial dependence present in the data. The limiting null distributions of these test statistics are shown to follow a chi-square distribution, and their asymptotic power under local alternatives is derived. A comprehensive set of simulation experiments illustrates their finite sample size and power properties. Funding: ESRC, ES/W000989/1

Direct Multi-Step Forecast based Evaluation of Nested Models via an Encompassing Test, December 2023 [under review]

Abstract: We introduce a novel approach for comparing out-of-sample multi-step forecasts obtained from a pair of nested models that is based on the forecast encompassing principle. Our proposed approach relies on an alternative way of testing the population moment restriction implied by the forecast encompassing principle and that links the forecast errors from the two competing models in a particular way. Its key advantage is that it is able to bypass the variance degeneracy problem afflicting model based forecast comparisons across nested models. It results in a test statistic whose limiting distribution is standard normal and which is particularly simple to construct. Multi-step forecasts are handled through two alternative ways. A standard approach that relies on a HAC type studentization of our test statistic and an alternative self-normalization based approach which bypasses the need to estimate long-run variances. All our inferences are shown to be able to accommodate a variety of predictor types including stationary, highly-persistent and purely deterministic predictors without affecting the way inferences are implemented.  Funding: ESRC, ES/W000989/1