An introduction to data registration in functional data analysis

The advancement of computation allows the amalgamation of high-dimensional data and large data sets at many different time points which needs meticulous research on functional data analysis and modeling. Data registration as one of the primary issues in exploratory functional data analysis involves transformations of its argument instead of the functions. On the other hand, insufficient attention to the analysis of registration data can result in critical protection gaps. In this paper, the shift method and nonlinear landmark registration methods will be discussed and an algorithm for data registration is presented. The applicability and advantages of data registration are investigated through a real data example.

کلیدواژه ها:

functional data analysis ، data registration ، shift registration method.