Periodicity Analysis of Most Time Series Methods: A Review
Muneera Yousif, Musab Ahmed Mohammed, Mete Çelik, Asmaa E. Hamdoon, Riham Jassim, Noor M. Abdullah · 2024
One of the most important methods of predicting the future is through past events and data repeated over time, as time series are those data indexed using time sequentially on data points distributed according to time, and periodic time series analysis is the analysis of a series of recorded measurements of a phenomenon that are repeated more than once during specific times periods. Periodicity detection has an impact on many areas in the world, such as climate, motion, education et cetera. It is important to consider in solving problems, moreover, there are many data sets in the world, the most prominent of which is time series data, and this type of data is constantly increasing. In the study, several data periodicity analysis methods (such as autocorrelation, wavelet transform, DTW, RNN, LSTM) were inspected.