Peer Review #1 of "Generalized relational tensors for chaotic time series (v0.1)"
2023
The paper deals with a generalized relational tensor, a novel discrete model to store information about a time series, and algorithms (1) to construct the model, (2) to generate a time series from the model, and (3) to predict a time series.The algorithms combine the concept of generalized z-vectors with ant colony optimization techniques.To estimate the quality of the storing/re-generating procedure, a difference between the characteristics of the initial and regenerated time series is used.For chaotic time series, a difference between characteristics of the initial time series (the largest Lyapunov exponent, the auto-correlation function) and those of the time series re-generated from a model is used to assess the effectiveness of the algorithms in question.The approach has shown fairly good results for periodic and benchmark chaotic time series and satisfactory results for real-world chaotic data.