3.1 Topics Covered

• Understand the mathematics behind classic metric multidimensional scaling (MDS)
• Understand the principle behind the SMACOF algorithm
• Carry out multidimensional scaling for a set of variables
• Gain insight into the various options used in MDS analysis
• Visualize and interpret the results of MDS
• Compare closeness in attribute space to closeness in geographic space
• Carry out local neighbor match test using MDS neighbors
• Implement density based clustering on MDS coordinates

GeoDa Functions

• Clusters > MDS
  • select variables
  • MDS methods
  • MDS parameters
  • saving MDS results
  • spatial weights from MDS results

Toolbar Icons

Figure 3.1: Clusters > PCA | MDS | t-SNE