Editing Representative sequences (section)

== Social sciences ==
[[File:Fg-rep-seq-biofam.png|thumb|Representative sequences covering 27% of 2000 cohabitation sequences between age 15 and 30 (extract of biographical data from the Swiss Household Panel)]]
In [[Sequence analysis in social sciences]], representative sequences are used to summarize sets of sequences describing for example the family life course or professional career of several thousands individuals.<ref name=":18">{{Cite journal |last1=Gabadinho |first1=Alexis |last2=Ritschard |first2=Gilbert |date=2013 |editor-last=Levy |editor-first=René |editor2-last=Widmer |editor2-first=Eric D. |title=Searching for typical life trajectories, applied to childbirth histories |url=https://www.researchgate.net/publication/287202533 |journal=Gendered Life Courses, Between Standardization and Individualization: A European Approach Applied to Switzerland |publisher=LIT |publication-place=Zurich |pages=287–312}}</ref>

The identification of representative sequences<ref name=":0">{{Citation |last1=Gabadinho |first1=Alexis |title=Extracting and Rendering Representative Sequences |date=2011 |url=http://link.springer.com/10.1007/978-3-642-19032-2_7 |work=Knowledge Discovery, Knowledge Engineering and Knowledge Management |volume=128 |pages=94–106 |editor-last=Fred |editor-first=Ana |access-date=2023-06-12 |place=Berlin, Heidelberg |publisher=Springer Berlin Heidelberg |doi=10.1007/978-3-642-19032-2_7 |isbn=978-3-642-19031-5 |last2=Ritschard |first2=Gilbert |last3=Studer |first3=Matthias |last4=Müller |first4=Nicolas S. |series=Communications in Computer and Information Science |editor2-last=Dietz |editor2-first=Jan L. G. |editor3-last=Liu |editor3-first=Kecheng |editor4-last=Filipe |editor4-first=Joaquim|url-access=subscription }}</ref><ref name=":18" /> proceeds from the pairwise dissimilarities between sequences. One typical solution is the medoid sequence, i.e., the observed sequence that minimizes the sum of its distances to all other sequences in the set. An other solution is the densest observed sequence, i.e., the sequence with the greatest number of other sequences in its neighborhood. When the diversity of the sequences is large, a single representative is often insufficient to efficiently characterize the set. In such cases, an as small as possible set of representative sequences covering (i.e., which includes in at least one neighborhood of a representative) a given percentage of all sequences is searched.

A solution also considered is to select the medoids of relative frequency groups. More specifically, the method consists in sorting the sequences (for example, according to the first principal coordinate of the pairwise dissimilarity matrix), splitting the sorted list into equal sized groups (called relative frequency groups), and selecting the medoids of the equal sized groups.<ref>{{Cite journal |last1=Fasang |first1=Anette Eva |last2=Liao |first2=Tim Futing |date=2014 |title=Visualizing Sequences in the Social Sciences: Relative Frequency Sequence Plots |url=http://journals.sagepub.com/doi/10.1177/0049124113506563 |journal=Sociological Methods & Research |language=en |volume=43 |issue=4 |pages=643–676 |doi=10.1177/0049124113506563 |issn=0049-1241 |s2cid=61487252 |hdl=10419/209702|hdl-access=free }}</ref>

The methods for identifying representative sequences described above have been implemented in the R package [https://cran.r-project.org/package=TraMineR TraMineR].<ref>{{Cite journal |last1=Gabadinho |first1=Alexis |last2=Ritschard |first2=Gilbert |last3=Müller |first3=Nicolas S. |last4=Studer |first4=Matthias |date=2011 |title=Analyzing and Visualizing State Sequences in R with TraMineR |url=http://www.jstatsoft.org/v40/i04/ |journal=Journal of Statistical Software |language=en |volume=40 |issue=4 |doi=10.18637/jss.v040.i04 |issn=1548-7660|doi-access=free }}</ref>