This is the authors' abstract. We don't add key points for this paper.
Autocorrelation is one of the basic concepts in time series analysis and attracts persistent attentions for more than one hundred years. Correlations in time series induce complicated patterns and their interdependent relationships. Hence, the autocorrelation has a network structure, in which the nodes are the patterns and the edges the interdependent relationships between the patterns. However, in the standard procedures for estimating autocorrelation the network structure is lost in the procedure of statistical average. In the present paper we propose a concept called local pattern interdependent (LPID) network to display the structure of the autocorrelation. It is found that the series generated with the theoretical models of the Autoregressive Process and the series of empirical records of stock market indices share several behaviors. Their differences come from the weights, among which the self-linkage of a specific pattern is significantly large and can take subsequently a measure to distinguish the series. Hence, the LPID network provides a unified framework to investigate time series generated by different mechanisms. What is more, the results tell us that this network structure of autocorrelation stores a kind of high-order correlation compared with the traditional definition of autocorrelation.
Answers are grounded in this paper and cite their sources.
No takes yet. Share an insight, caveat, or question.
Ren et al. (2025) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: