@article{
 title = {On the number of principal components: A test of dimensionality based on measurements of similarity between matrices},
 type = {article},
 year = {2008},
 pages = {2228-2237},
 volume = {52},
 id = {d4cf9ad3-8bff-335b-b4cf-658153d424bc},
 created = {2010-11-03T21:13:25.000Z},
 file_attached = {true},
 profile_id = {976aa121-3316-304c-8340-7ca54d70abe6},
 last_modified = {2017-03-16T14:38:37.564Z},
 read = {true},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Dray2008b},
 source_type = {article},
 private_publication = {false},
 abstract = {An important problem in principal component analysis (PCA) is the
estimation of the correct number of components to retain. PCA is
most often used to reduce a set of observed variables to a new set
of variables of lower dimensionality. The choice of this dimensionality
is a crucial step for the interpretation of results or subsequent
analyses, because it could lead to a loss of information (underestimation)
or the introduction of random noise (overestimation). New techniques
are proposed to evaluate the dimensionality in PCA. They are based
on similarity measurements, singular value decomposition and permutation
procedures. A simulation study is conducted to evaluate the relative
merits of the proposed approaches. Results showed that one method
based on the RV coefficient is very accurate and seems to be more
efficient than other existing approaches.},
 bibtype = {article},
 author = {Dray, Stéphane},
 journal = {Computational Statistics and Data Analysis}
}

{"_id":"RnB8gvRMf8m5XDsJT","bibbaseid":"dray-onthenumberofprincipalcomponentsatestofdimensionalitybasedonmeasurementsofsimilaritybetweenmatrices-2008","author_short":["Dray, S."],"bibdata":{"title":"On the number of principal components: A test of dimensionality based on measurements of similarity between matrices","type":"article","year":"2008","pages":"2228-2237","volume":"52","id":"d4cf9ad3-8bff-335b-b4cf-658153d424bc","created":"2010-11-03T21:13:25.000Z","file_attached":"true","profile_id":"976aa121-3316-304c-8340-7ca54d70abe6","last_modified":"2017-03-16T14:38:37.564Z","read":"true","starred":false,"authored":"true","confirmed":"true","hidden":false,"citation_key":"Dray2008b","source_type":"article","private_publication":false,"abstract":"An important problem in principal component analysis (PCA) is the\nestimation of the correct number of components to retain. PCA is\nmost often used to reduce a set of observed variables to a new set\nof variables of lower dimensionality. The choice of this dimensionality\nis a crucial step for the interpretation of results or subsequent\nanalyses, because it could lead to a loss of information (underestimation)\nor the introduction of random noise (overestimation). New techniques\nare proposed to evaluate the dimensionality in PCA. They are based\non similarity measurements, singular value decomposition and permutation\nprocedures. A simulation study is conducted to evaluate the relative\nmerits of the proposed approaches. Results showed that one method\nbased on the RV coefficient is very accurate and seems to be more\nefficient than other existing approaches.","bibtype":"article","author":"Dray, Stéphane","journal":"Computational Statistics and Data Analysis","bibtex":"@article{\n title = {On the number of principal components: A test of dimensionality based on measurements of similarity between matrices},\n type = {article},\n year = {2008},\n pages = {2228-2237},\n volume = {52},\n id = {d4cf9ad3-8bff-335b-b4cf-658153d424bc},\n created = {2010-11-03T21:13:25.000Z},\n file_attached = {true},\n profile_id = {976aa121-3316-304c-8340-7ca54d70abe6},\n last_modified = {2017-03-16T14:38:37.564Z},\n read = {true},\n starred = {false},\n authored = {true},\n confirmed = {true},\n hidden = {false},\n citation_key = {Dray2008b},\n source_type = {article},\n private_publication = {false},\n abstract = {An important problem in principal component analysis (PCA) is the\nestimation of the correct number of components to retain. PCA is\nmost often used to reduce a set of observed variables to a new set\nof variables of lower dimensionality. The choice of this dimensionality\nis a crucial step for the interpretation of results or subsequent\nanalyses, because it could lead to a loss of information (underestimation)\nor the introduction of random noise (overestimation). New techniques\nare proposed to evaluate the dimensionality in PCA. They are based\non similarity measurements, singular value decomposition and permutation\nprocedures. A simulation study is conducted to evaluate the relative\nmerits of the proposed approaches. Results showed that one method\nbased on the RV coefficient is very accurate and seems to be more\nefficient than other existing approaches.},\n bibtype = {article},\n author = {Dray, Stéphane},\n journal = {Computational Statistics and Data Analysis}\n}","author_short":["Dray, S."],"biburl":"https://bibbase.org/service/mendeley/2845861","bibbaseid":"dray-onthenumberofprincipalcomponentsatestofdimensionalitybasedonmeasurementsofsimilaritybetweenmatrices-2008","role":"author","urls":{},"metadata":{"authorlinks":{}},"downloads":0},"bibtype":"article","biburl":"https://bibbase.org/service/mendeley/2845861","dataSources":["2252seNhipfTmjEBQ"],"keywords":[],"search_terms":["number","principal","components","test","dimensionality","based","measurements","similarity","between","matrices","dray"],"title":"On the number of principal components: A test of dimensionality based on measurements of similarity between matrices","year":2008}