Persistence, seasonality, and fractional integration within a nonlinear framework: Evidence from US citizens’ overseas travel

This research note examines data on US citizens’ overseas air passenger travel with respect to the degree of persistence, seasonality, nonlinearities, and fractional integration. Based on seasonally differenced data with allowance for autocorrelated errors, we find evidence of nonlinearity with moderate persistence and mean reversion in US overseas air travel. The results suggest that shocks to US overseas air travel will dissipate over time, and as such, the observed persistence and nonlinearity can provide useful information in the modeling and forecasting of future behavior.

The Effect of Spatiotemporal Resolution Degradation on the Accuracy of IMERG Products over the Huai River Basin

The rapid development of the Integrated Multisatellite Retrievals for Global Precipitation Measurement (IMERG) precipitation product provides new opportunities for a wide range of Earth system and natural hazard applications. Spatiotemporal averaging is a common method for IMERG users to acquire suitable resolutions specific to their research or application purpose and has a direct impact on the overall quality of IMERG precipitation estimates. Here, three different IMERG, version 06 (V06), latency run products (i.e., early, late, and final) are assessed against a ground-based benchmark along a continuous series of spatiotemporal resolutions over the Huai River basin (HuaiRB) between June 2014 and May 2017. In general, IMERG products better capture the spatial pattern of precipitation, and demonstrate better reliability, in the southern portion of the HuaiRB relative to its northern region. Furthermore, the degradation of spatiotemporal resolution is associated with better rain/no-rain determination and the consistent improvement of rainfall product performance metrics. This improvement is more pronounced for IMERG products at fine spatiotemporal resolution. However, due to the presence of autocorrelated errors, the performance improvement associated with the degradation of spatiotemporal resolution is less than theoretical expectations assuming purely uncorrelated errors. Component analysis indicates that while both temporal and spatial aggregation do not mitigate temporally autocorrelated errors, temporal averaging can remove spatially autocorrelated error. Hence, temporal averaging is found to be more effective than spatial averaging for improving the quality of IMERG products. These results will inform users of the reliability of IMERG products at different spatiotemporal scales and assist in unifying former disparate validation assessments applied at different scales within the literature.

Statistical Inference for Least Absolute Deviation Regression with Autocorrelated Errors

The method of least absolute deviation provides a robust alternative to least squares, particularly when the data follow distributions that are non-normal and subject to outliers. While inference in least squares estimation is well understood, inferential procedures in the situation of least absolute deviation estimation have not been studied as extensively, particularly in the presence of autocorrelation. In this search, we study two alternative significance test procedures in least absolute deviation regression, along with two approaches used to correct for serial correlation. The study is based on a Monte Carlo simulation, and comparisons are made based on observed significance levels.