2003
DOI: 10.1111/1467-842x.00285
|View full text |Cite
|
Sign up to set email alerts
|

Generalized discriminant analysis based on distances

Abstract: This paper describes a method of generalized discriminant analysis based on a dissimilarity matrix to test for differences in a priori groups of multivariate observations. Use of classical multidimensional scaling produces a low-dimensional representation of the data for which Euclidean distances approximate the original dissimilarities. The resulting scores are then analysed using discriminant analysis, giving tests based on the canonical correlations. The asymptotic distributions of these statistics under pe… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

4
369
0
2

Year Published

2010
2010
2020
2020

Publication Types

Select...
8
2

Relationship

0
10

Authors

Journals

citations
Cited by 644 publications
(390 citation statements)
references
References 22 publications
4
369
0
2
Order By: Relevance
“…In case of significant 'shelf-slope  year' interactions, pair-wise tests of 'year' within 'shelf-slope' were performed to investigate the difference between years at shelf or slope location. Because of the restricted number of possible permutations in 'shelf-slope' effect and pair-wise tests, p-values were obtained from Monte Carlo samplings (Anderson and Robinson, 2003). A Euclidean distance-based resemblance matrix was used and probability values were obtained by permutation (n = 10 4 ).…”
Section: Statistical Treatment Of Datamentioning
confidence: 99%
“…In case of significant 'shelf-slope  year' interactions, pair-wise tests of 'year' within 'shelf-slope' were performed to investigate the difference between years at shelf or slope location. Because of the restricted number of possible permutations in 'shelf-slope' effect and pair-wise tests, p-values were obtained from Monte Carlo samplings (Anderson and Robinson, 2003). A Euclidean distance-based resemblance matrix was used and probability values were obtained by permutation (n = 10 4 ).…”
Section: Statistical Treatment Of Datamentioning
confidence: 99%
“…PERMANOVA partitions the total sum of squares based on the full experimental design (given below), and calculates a distance based pseudo-F statistic for each term in the model, based on the expectations of the mean squares. P-values are obtained using a permutation procedure for each term (Anderson et al, 2008), or via Monte Carlo random drawn from an asymptotic permutation distribution if too few permutations are available for a given test (Anderson and Robinson, 2003).…”
Section: Data Analysesmentioning
confidence: 99%
“…Pair-wise tests between stations obtained through the PERMANOVA analysis on one-way experimental design are reported in Table 3. As the number of unique values under permutations was very low, P-values were obtained using Monte Carlo samples from the asymptotic permutation distribution (Anderson and Robinson, 2003). Benthic species richness and diversity showed variability between stations.…”
Section: Resultsmentioning
confidence: 99%