Options for factorial analysis of mixed data in XLSTAT Sometimes, only the first and second components are necessary to explain a large percentage of the variance and so you will be able analyze a two-dimensional projection of what was initially dozens of variables. The observations and the variables end up being represented as points in orthogonal two-dimensional spaces. This is when the dimensional reduction occurs because a small number of components will be enough to explain a high percentage of the variance. The number of principal components is chosen depending on the explained percentage of variance of the model by each component. To do so, it does a series of statistical transformations, including calculations of the correlation matrix, eigenvalues and eigenvectors, on a set of qualitative and/or quantitative variables in order to project them on a vector space generated by orthogonal components. Similarly to other factorial analysis methods, PCAmix aims to reduce data dimensionality as well as to identify nearness between variables but also proximity between the observations. This method can be seen as a mixture of two popular methods of factorial analysis: Principal Component Analysis (PCA) which allows to study an observations/quantitative variables table and Multiple Correspondence Analysis (MCA) which allows to study an observations/qualitative variables table. The method used in Xlstat is called PCAmix and was developed by Chavent et al (2014). A few variants of this method have been developed since then (Escofier 1979, Pagès 2004). What is Factorial analysis of mixed data?įactorial analysis of mixed data is a method initially developed by Hill and Smith (1972).
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |