Variable Selection in Principal Component Analysis

9. Multidimensional Data Analysis

“…So it is desirable to supply some system to ordinary users so that they can apply some of the approaches to their own data sets with any of the criteria/methods or can get a suggestion based on the results of the approaches. This leads to a plan to implement a way to perform two approaches or provide the information summarizing the results of them into the statistical package "VASPCA" for variable selection in PCA (Mori,1997;Mori et al, 2000a).…”

“…Then those who want to select a subset of variables are advised to apply a suitable method for their purpose of selection and/or to try some methods and select one by comparing the results. To support them and provide tools for variable selection in PCA, web sites have been developed, in which anyone can easily perform variable selection in PCA with any existing criteria (e.g., Mori et al (2000a) and see URLs in Appendix).…”

Variable Neighborhood Search Heuristics for Selecting a Subset of Variables in Principal Component Analysis

“…The forward selection and backward elimination algorithms can be modified to permit both the addition and deletion of variables during the subset selection process (Cadima et al, 2004;Iizuka, Mori, Tarumi, & Tanaka, 2003;Mori, Iizuka, Tarumi, & Tanaka, 2007). These modified algorithms are known as forward stepwise and backward stepwise heuristics.…”

Acceleration of Convergence of the Alternating Least Squares Algorithm for Nonlinear Principal Components Analysis

“…In order to find a subset of q variables, we employ Backward elimination and Forward selection of Mori et al [Mori et al, 1998;Mori et al, 2006] as cost-saving stepwise selection procedures in which only one variable is removed or added sequentially.…”