TraMineR: a Sequence Analysis Toolkit
TraMineR is a
R-package for manipulating, describing, visualizing, and analyzing
sequences of states or events, and more generally discrete sequence data. Its primary aim is the analysis of longitudinal data in the social sciences, such as data describing careers, family trajectories, and time-use. However, most of its features also apply to many other kinds of categorical sequence data. They include:
- Handling of longitudinal data and conversion between various sequence formats
- Plotting sequences (density plot, frequency plot, index plot and more)
- Individual longitudinal characteristics of sequences (length, time in each state, longitudinal entropy, turbulence, complexity, insecurity, and many more)
- Sequence of cross-sectional characteristics by time point (transversal state distribution, transversal entropy, modal state)
- Other aggregated characteristics (transition rates, average duration in each state, sequence frequency)
- Dissimilarities between pairs of sequences: Edit distances (Optimal matching and many variants of it, Hamming, ...), Metrics based on counts of common attributes (based on longest common subsequence, on number of matching subsequences, ...), Distances between state distributions (Chi-squared, Euclidean, ...)
- Multidomain/multichannel sequence analysis (multichannel distances, association between domains, multichannel plots)
- Medoid and heterogeneity measure of a set of sequences
- Discovering and plotting representative sequences
- ANOVA-like analysis of sequences
- Regression trees of sequences
- Parallel coordinate plot of state and event sequences
- Extracting frequent event subsequences
- Identifying most discriminating event subsequences
- Association rules between subsequences
Click
here for a short preview of what TraMineR can do for you!
What does TraMineR stand for?
It is a contraction of Life Trajectory Miner for R (and was inspired by the authors' taste for Gewürztraminer wine).
Who is developing TraMineR?
TraMineR is developed at
the
Institute of Demography and Socioeconomics (IDESO),
University of Geneva, Switzerland. The package is currently maintained by
TraMineR was originally created by
Alexis Gabadinho,
Gilbert Ritschard,
Matthias Studer, and Nicolas S. Müller within the project
Mining event histories funded by the
Swiss National Foundation for Scientific Research under grants FN-116416 and FN-122230. It later benefited from contributions of
Reto Bürgin and
Pierre-Alexandre Fonta thanks to support of the
NCCR LIVES - overcoming vulnerability: life course perspectives.
Other R packages from the TraMineR team
-
TraMineRextras, TraMineR ancillary functions
-
PST, Alexis Gabadinho's package for Probabilistic Suffix Trees
-
WeightedCluster, Matthias Studer's clustering package
-
seqSHP, building sequences from SHP waves
-
vcrpart, Reto Bürgin's package for Tree-based varying coefficients for ordinal mixed regression models and generalized linear regression models.
Useful links
