2 credit corresponds to 15h/class
OBJECTIVES OF THE DISCIPLINE:
Familiarize students with the basic concepts and techniques of multivariate analysis. Give an overview of multivariate data analysis methods and show their applications to different chemical problems.
SUMMARY:
1. Introduction
2. Multivariate Data
3. Principal Component Analysis
4. Calibration and Classification
5. Cluster Analysis
ANALYTICAL PROGRAM:
1. Introduction
the. Definition and History
b. Examples
2. Multivariate Data
the. Definitions
b. Processing
w. Covariance and Correlation
d. Distances and Similarities
and. Identification of Outliers
f. Latent Linear Variables
3. Principal Component Analysis
the. Concepts
b. Algorithms
w. Assessment and Diagnostics
d. Complementary Methods
and. Examples
4. Calibration
the. Concepts
b. Regression Models
w. Simple Linear Regression
d. Robust Regression
and. Variable Selection
f. Regression by Principal Components
g. Partial Least Squares Regression
h. Examples
5. Classification
the. Concepts
b. Linear Classification Methods
w. Kernel and Prototype-based Methods
d. Classification Trees
and. Neural Networks
f. Support Vector Machine
g. Assessment
h. Examples
6. Cluster Analysis
the. Concepts
b. Distance and Similarity Measures
w. Partitioning Methods
d. Hierarchical Cluster Analysis Methods
and. Fuzzy Clustering
f. Model-Based Groupings
g. Trend Validation and Measurements
h. Examples
Bibliography
1. Introduction to multivariate statistical analysis in chemometrics / Kurt Varmuza and Peter Filzmoser, 2009.
2. Applied Multivariate Statistical Analysis / Wolfgang Härdle and Léopold Simar, 2003.
3. Chemometrics – Data Analysis for the Laboratory and Chemical Plant / Richard G. Brereton, 2003.
