Authors' reply to the discussion of 'Automatic change-point detection in time series via deep learning' at the discussion meeting on 'Probabilistic and statistical aspects of machine learning'
Li, J., Fearnhead, P., Fryzlewicz, P.
& Wang, T.
(2024).
Authors' reply to the discussion of 'Automatic change-point detection in time series via deep learning' at the discussion meeting on 'Probabilistic and statistical aspects of machine learning'.
Journal of the Royal Statistical Society. Series B: Statistical Methodology,
86(2), 332 - 334.
https://doi.org/10.1093/jrsssb/qkae008
& Wang, T.
(2024).
Authors' reply to the discussion of 'Automatic change-point detection in time series via deep learning' at the discussion meeting on 'Probabilistic and statistical aspects of machine learning'.
Journal of the Royal Statistical Society. Series B: Statistical Methodology,
86(2), 332 - 334.
https://doi.org/10.1093/jrsssb/qkae008
We would like to thank the proposer, seconder, and all discussants for their time in reading our article and their thought-provoking comments. We are glad to find a broad consensus that neural-network-based approach offers a flexible framework for automatic change-point analysis. There are a number of common themes to the comments, and we have therefore structured our response around the topics of the theory, training, the importance of standardization and possible extensions, before addressing some of the remaining individual comments.
