Modeling Data Based on the Correlations - Lecture Slides | STAT 530, Study notes of Statistics

Download Modeling Data Based on the Correlations - Lecture Slides | STAT 530 and more Study notes Statistics in PDF only on Docsity! 1 STAT 530/J530 B.Habing Univ. of S.C. 1 STAT 530/J530 October 11th, 2005 Instructor: Brian Habing Department of Statistics LeConte 203 Telephone: 803-777-3578 E-mail: habing@stat.sc.edu STAT 530/J530 B.Habing Univ. of S.C. 2 So Far... The Fundamentals • Introduction to the R, SAS, and SPSS • Mutlivariate Data • Multivariate Normal Distribution • Statistical Graphics STAT 530/J530 B.Habing Univ. of S.C. 3 So Far... Methods Based for Simplifying or Modeling Data based on Correlations or Covariances • Principal Components • Factor Analysis 2 STAT 530/J530 B.Habing Univ. of S.C. 4 Next Methods Based for Representing of Finding Patterns in Data Based on Distances • Multidimensional Scaling and Correspondence Analysis • Cluster Analysis STAT 530/J530 B.Habing Univ. of S.C. 5 Distances A distance (or dissimilarity) dij between two points i and j must satisfy: 1) Symmetry: dij = dji 2) Non-negativity: dij ≥ 0 3) Identification: dii = 0 STAT 530/J530 B.Habing Univ. of S.C. 6 Metric A distance dij is called a metric if it also satisfies: 4) Definiteness: dij = 0 if and only if i =j. 5) Triangle Inequality: dij ≤ dik+ djk