Assessing Dissimilarity Between Multivariate Time Series
Computes sum of distances between consecutive samples in a multivariat...
Computes a multivariate distance between two vectors.
Computes distance matrices among the samples of two or more multivaria...
Computes distance among pairs of aligned samples in two or more multiv...
distantia: Assessing Dissimilarity Between Multivariate Time Series
Formats the output of psi.
Handles emtpy and NA data in a multivariate time series.
Extracts the least cost from a least-cost path.
Computes a least cost matrix from a distance matrix.
Find the least cost path in a least cost matrix.
Extracts the least-cost from a least cost matrix by trimming blocks.
Plots distance matrices and least cost paths.
Prepare sequences for a comparison analysis.
Computes sum of distances between consecutive samples in a multivariat...
Computes the contribution to dissimilarity of each variable.
Computes the contribution to dissimilarity of each variable using work...
Computes the dissimilarity measure psi on restricted permutations ...
Computes the dissimilarity measure psi on restricted permutations ...
Finds the section in a long sequence that better matches a short seque...
Computes the dissimilarity measure psi on two or more sequences.
A refactored version of workflowPsi with a higher performance (hence...
Slots two sequences into a single composite sequence.
Transfers an attribute (time, age, depth) from one sequence to another
Provides tools to assess the dissimilarity between multivariate time-series. It is based on the psi measure described by Birks and Gordon (1985 <doi:10.1002/jqs.3390020110>), which computes dissimilarity between irregular time-series constrained by sample order. However, in this package the original idea has been extended to work with any kind of multivariate time-series, no matter whether they are regular, irregular, aligned or unaligned. Furthermore, the package allows to assess the significance of dissimilarity values by applying a restricted permutation test, allows to measure the contribution of individual variables to dissimilarity, and offers tools to transfer attributes (generally time or age, but other are possible) between sequences based on the similarity of their samples.