The program, written in R, is a modification of the R function acf. It plots the sample autocorrelations, the standard significance bounds in blue dotted lines and generalized significant bounds in red lines. Sample autocorrelations quite outside the bounds are generally considered as evidence against the white noise hypothesis. The standard significant bounds are 95% bounds for the sample autocorrelations of a strong (i.e. iid) white noise. The red bounds are valid for more general white noises, which can exhibit conditional heteroscedasticity or more general forms of nonlinear dynamics (see Francq, C. and Zakoïan, J-M. Bartlett's formula for a general class of non linear processes Journal of Time Series Analysis, 30, 449-465, 2009).
Oxford Bulletin of Economics and Statistics (2022)
Matsuoka Hideaki