This package provides general tools for analyzing non-Gaussian nonlinear multivariate time series models. The algorithm is described in the paper Nonlinear Time Series Modeling by LPTime, Nonparametric Empirical Learning., by Mukhopadhyay and Parzen (2013). The central idea behind LPTime time series modelling algorithm is to convert the original univariate time series X(t) into
Accepts possibly non-Gaussian non-linear univariate (stationary) time series data; converts it to multivariate LP-transformed series and fits a vector autoregressive (VAR) model.
Evaluates m LP moments of a random variable. Estimates LP-comoment matrix of order m \times m between X and Y , i.e., covariance between the LP transformations of X and Y; where the random variables could be discrete or continuous.