@article{fdi:010074791,
  title = {{M}ultidimensional scaling with very large datasets},
  author = {{P}aradis, {E}mmanuel},
  editor = {},
  language = {{ENG}},
  abstract = {{M}ultidimensional scaling has a wide range of applications when observations are not continuous but it is possible to define a distance (or dissimilarity) among them. {H}owever, standard implementations are limited when analyzing very large datasets because they rely on eigendecomposition of the full distance matrix and require very long computing times and large quantities of memory. {H}ere, a new approach is developed based on projection of the observations in a space defined by a subset of the full dataset. {T}he method is easily implemented. {A} simulation study showed that its performance are satisfactory in different situations and can be run in a short time when the standard method takes a very long time or cannot be run because of memory requirements.},
  keywords = {{D}imension reduction ; {D}istance data ; {P}rojection method ; {R}andom matrices},
  booktitle = {},
  journal = {{J}ournal of {C}omputational and {G}raphical {S}tatistics},
  volume = {27},
  numero = {4},
  pages = {935--939},
  ISSN = {1061-8600},
  year = {2018},
  DOI = {10.1080/10618600.2018.1470001},
  URL = {https://www.documentation.ird.fr/hor/fdi:010074791},
}