Data Analysis

Objectives

Course objectives:

Knowledge and understanding of main techniques for Multivariate Data Analysis.
Presentation of numerous applications where univariate, bivariate and multivariate analysis associated to data with quantitative variables or qualitative variables, or both, are developed.
Use of MS Excel and SAS for statistical multivariate real data treatment.

General characterization

Code

100003

Credits

6.0

Responsible teacher

Frederico Miguel Campos Cruz Ribeiro de Jesus

Hours

Weekly - Available soon

Total - Available soon

Teaching language

Portuguese. If there are Erasmus students, classes will be taught in English

Prerequisites

n/a

Bibliography

Sharma, S. (1996). Applied Multivariate Techniques. New York, John Wiley & Sons, Inc.
Hair, J. F. (2010). Multivariate Data Analysis, Prentice Hall.
Reis, E. (2001). Estatística Multivariada Aplicada, Edições Silabo.
Branco, João, (2004) ¿ Uma Introdução à análise de clusters, Ed. Sociedade Portuguesa de Estatística
Course´s slides.

Teaching method

The curricular unit is based on mix of theoretical lectures and practical classes. Each session will introduce new concepts and methodologies, as well as the applications of the learnt concepts using different computational tools. Different learning strategies will be used, such as lectures, slide show demonstrations, step-by-step tutorials on how to approach practical examples, questions, and answers.

The practical component is focused in exploring the different computational tools by the students, including a discussion on the best approach under different scenarios.

Evaluation method

1^st Term: 1^st Test (20%) + 2^nd Test (35%) + Group Project (35%) + Participation (10%)

2^nd Term: 2^nd term Exam (60%) + Group Project (40%)

Notes:

1^st test purposed for October 17^that 11am (online); 2^nd test in 1^st term exam date.
Students are strongly encouraged to follow the 1^st term assessment;
Project¿s presentations will take place on December 16^th (09h30m ¿ 18h30m in 3 slots); Groups need to have exactly 4 members.
Minimum grade for tests/exam and group project is eight out of 20 points.

Subject matter

Introduction and Fundamentals of Data Analysis
1. Measurement Scales
2. Classification of Data Analysis´ Methods
3. Geometric Concepts of Data Manipulation
4. Fundamentals of Data Manipulation
Principal Components Analysis (PCA)
1. Geometry of PCA
2. Analytical Approach
3. Application of PCA in SAS
4. Issues relating to the use of PCA
Factor Analysis (FA)
1. Basic Concepts and Terminology of FA
2. Objectives of FA
3. Geometric view of FA
4. FA techniques
5. Application of PCA in SAS
6. FA versus PCA
Correspondence Factorial Analysis (CFA)
1. Basic Concepts and Terminology
2. Contingency Tables
3. Chi-Square Statistic and Inertia
4. CFA Interpretation
5. CFA Application
6. Multiple Correspondence Analysis
Cluster Analysis (CA)
1. Geometrical view of CA
2. Objectives of CA
3. Similarity measures
4. Hierarchical CA
5. Nonhierarchical CA
6. Application of CA in SAS
7. Similarity Measures
8. Reliability and External Validity of CA
9. Considerations about multiple methods of CA
Multidimensional Scalling (MDS)
1. Introduction and Objectives of MDS
2. Research Design
3. Assumptions
4. Deriving MDS Results and Overall Fit
5. Interpretation
6. Validation
7. MDS using SAS
Discriminant Analysis (DA)
1. Geometric view of DA
2. Analytical Approach to DA
3. Regression Approach to DA
4. Assumptions
5. External Validation of the Discriminant Function
Data Analysis Applications
1. Three-way PCA
2. Cluster Analysis using PCA or FA
3. Combining hierarchical and nonhierarchical CA
4. Some examples (Papers, studies, etc.)

Programs

Programs where the course is taught:

Bachelor Degree in Information Management