R bloggers

TSrepr – Time Series Representations in R

FavoriteLoadingAdd to favorites

I’m happy to announce a new package that has recently appeared on CRAN, called “TSrepr” (version 1.0.0: https://CRAN.R-project.org/package=TSrepr). The TSrepr package contains methods of time series representations (dimensionality reduction, feature extraction or preprocessing) and several other useful helper methods and functions. Time series representation can be defined as follows: Let ( x ) be a time series of length ( n ), then representation of ( x ) is a model ( hat{x} ) with reduced dimensionality ( p ) ( (p < n) ) such that ( hat{x} ) approximates closely ( x ) (Esling and Agon (2012)). Time series representations are used for: significant reduction of the time series dimensionality emphasis on fundamental (essential) shape characteristics implicit noise handling reducing the dimension will reduce memory requirements and computational complexity of consequent machine learning methods. Therefore, they are awesome!…
Original Post: TSrepr – Time Series Representations in R