数値実験の比較による, 関係類似度を用いたRsn法の利点 [in Japanese] Advantages of Rsn Method over Correspondence Analysis Used for Simulated Data [in Japanese]
Access this Article
Search this Article
In order to indicate the advantages of the Rsn method, we compared it with the widely used method, correspondence analysis (CA) based on the data simulated to those of benthic macrofauna. It is proved that the Rsn method has the following advantages. The sites of similar species composition are displayed in a mass, and their patterns correspond well with the spatial distributions of the species, that is preferable to the cluster analysis. The cumulative ratio of eigen values of the Rsn matrix is relatively high. This means that an error component included in the matrix is low, while the essential information is well preserved. The Rsn method, therefore, is applicable to the data analysis of benthic macrofauna.
NSUGAF 64(1), 56-64, 1998-01-15
The Japanese Society of Fisheries Science