Main Content

Multivariate Analysis

Multivariate Analysis (MVA) refers to statistical techniques used to analyze data arising from many variables. If information is organized in data tables containing rows and columns, MVA can help to process the information in a meaningful way.

MVA can be mainly used to:

  • Find local correlations between variables/outputs (main and interactions effect analysis don’t give this information!)
  • Have an immediate visualisation of the response of the system at a particular combinations of the variables (visualise multidimensional dataset)
  • Find and analyse local groups of data having similar instances (similar response of the system)
     
Image Self Organizing Map (SOM)

Self Organizing Map (SOM)

Project a multi-dimensional space onto a two-dimensional map.  Learn more..

Image Hierarchical Clustering

Hierarchical Clustering

Group designs into clusters with a bottom-up strategy. Learn more..

Image Partitive Clustering

Partitive Clustering

Divide designs space into "k" clusters with an iterative strategy. Learn more..

Image Principal Component Analysis (PCA)

Principal Component Analysis (PCA)

Project a multi-dimensional space onto the directions of maximum variability.  Learn more..

Image Multi-dimensional Scaling

Multi-dimensional Scaling

Project a multi-dimensional space preserving the inter-point distances.  Learn more..

DOWNLOAD ► PDF (Use of Multi-Variate-Data-Analysis Techniques in modeFRONTIER for Efficient Optimization and Decision Making)

NEWSLETTER - sign up!

description
follow us on Follow esteco_mF on Twitter