CMStatistics 2017: Start Registration
View Submission - CFE
Title: Testing for autocorrelation in non-linear and non-stationary regressions Authors:  Vanessa Berenguer Rico - University of Oxford (United Kingdom) [presenting]
Yingyu Guo - University of Oxford (United Kingdom)
Bent Nielsen - University of Oxford (United Kingdom)
Abstract: An auxiliary-regression-test is analyzed for autocorrelation of residuals in non-linear and non-stationary regressions. It is shown that the test statistic is asymptotically chi-square distributed under the null of no autocorrelation. Monte Carlo simulations reveal that the test statistic has good finite sample properties.