Part I
Course Duration: One Semester
Credit Units: 3
Level: B4
Medium of Instruction: English
Prerequisites: Nil
Precursors: MS3215 Linear Models
Equivalent Courses: Nil
Exclusive Courses: Nil
Part II
Course Aims:
develop the ability in students to analyse real life multivariate data.
Course Intended Learning Outcomes (CILOs)
Upon successful completion of this course, students should be able to:
1. apply basic concepts and techniques of commonly used multivariate statistical methods;
2. select appropriate techniques to solve multivariate data problems;
3. employ SAS computer software for implementing multivariate techniques.
Teaching and learning Activities (TLAs)
(Indicative of likely activities and tasks designed to facilitate students’ achievement of the CILOs. Final details will be provided to students in their first week of attendance in this course)
Including:
Lecture
Explain basic concepts and techniques of commonly used multivariate methods.
Computer Laboratory
Demonstration in the use of SAS computer software and discussion of computer exercises and problem sets.
Projects
Students will form teams or work individually on solving multivariate data problems and submit written reports. One of the projects may involve analysing business survey data that are collected in their Business Survey Design course.
Constructive Alignment of CILOs and Teaching and Learning Activities
| Teaching and Learning Activity |
CILO | Lecture | Computer Laboratory | Projects |
1 | Yes | Yes | Yes |
2 | Yes | Yes | Yes |
3 | Yes | Yes | Yes |
Assessment Tasks/Activities
(Indicative of likely activities and tasks designed to assess how well the students achieve the CILOs. Final details will be provided to students in their first week of attendance in this course)
Written examination (2 hours) | 50% |
Test | 20% |
Project report | 30% |
Total | 100% |
Constructive Alignment of CILOs and Assessment Methods
CILO | Written Examination
(2 hours) | Test | Project Reports |
1 | Yes | Yes | Yes |
2 | Yes | Yes | Yes |
3 | Yes | Yes | Yes |
Assessment Weights on CILOs and Assessment Methods
CILO
| Assessment Method |
Row Total
|
Written Examination
(2 hours) | Test | Project
Reports |
1 | 30 | 12 | 9 | 51 |
2 | 10 | 4 | 6 | 20 |
3 | 10 | 4 | 15 | 29 |
Column Total | 50 | 20 | 30 | 100 |
Grading of Student Achievement:
Written Examination and Test
Letter Grade | Grade Point | Grade Definitions | |
A+
A
A- | 4.3
4.0
3.7 | Excellent: | Strong evidence of original thinking; good organization, capacity to analyse and synthesize; superior application of subject matter; evidence of extensive knowledge base. |
B+
B
B- | 3.3
3.0
2.7 | Good: | Evidence of being able to apply the subject matter; evidence of critical capacity and analytic ability; reasonable understanding of issues; relevant use of literature. |
C+
C
C- | 2.3
2.0
1.7 | Adequate: | Some evidence of being able to apply the subject matter; some evidence of critical capacity and analytic ability; some evidence of understanding the issues; ability to develop solutions to simple problems. |
D | 1.0 | Marginal: | Sufficient ability to apply the subject matter to enable the student to progress without repeating the course. |
F | 0.0 | Failure: | Little or no evidence of being able to apply the subject matter; weakness in critical and analytic skills; limited or irrelevant use of literature. |
Project Reports
Letter Grade | Grade Point | Grade Definitions | |
A+
A
A- | 4.3
4.0
3.7 | Excellent: | Strong evidence of original thinking; good organization, capacity to analyse and synthesize; superior application of subject matter; evidence of extensive knowledge base. Highly effective use of language and excellent presentation skills. |
B+
B
B- | 3.3
3.0
2.7 | Good: | Evidence of being able to apply the subject matter; evidence of critical capacity and analytic ability; reasonable understanding of issues; relevant use of literature. Effective use of language and good presentation skills. |
C+
C
C- | 2.3
2.0
1.7 | Adequate: | Some evidence of being able to apply the subject matter; some evidence of critical capacity and analytic ability; some evidence of understanding the issues; ability to develop solutions to simple problems. Adequate command of the language and presentation skills. |
D | 1.0 | Marginal: | Sufficient ability to apply the subject matter to enable the student to progress without repeating the project. Inadequate command of the language and little presentation skills. |
F | 0.0 | Failure: | Little or no evidence of being able to apply the subject matter; weakness in critical and analytic skills; limited or irrelevant use of literature. Poor use of the language and presentation skills. |
Part III
Keyword Syllabus:
Introduction
An overview of multivariate methods. The structure of data matrix. Summary statistics for populations and samples. Random vector. Linear combination of random variables.
Multivariate Normal Distribution
Multivariate normal distribution and its properties. Methods of checking multivariate normal distribution assumption. Inference about mean vectors.
Multivariate Analysis of Variance
One-way multivariate analysis of variance (MANOVA).
Principal Components Analysis
Basic concepts of principal components. Estimation of principal components. Determining the number of principal components. How to perform principal components analysis using computer package. Interpretation of computer output. Business applications such as index construction.
Factor Analysis
Basic concepts of factor analysis. Methods of parameter estimation. Orthogonal and oblique rotations of factors. Estimation of factor scores. How to perform factor analysis using computer package. Interpretation of computer output. Factor analysis versus principal components analysis. Business applications such as attitude measurement.
Discriminant Analysis
Fisher's linear discriminant function. Optimal classification rules for two multivariate populations. Expected cost and total probability of misclassification. Quadratic discriminant function. Evaluation of classification functions. Discrimination among several populations. Stepwise discriminant analysis. How to perform discriminant analysis using computer package. Interpretation of computer output. Business applications such as credit analysis, bankruptcy prediction.
Cluster Analysis
Similarly measures. Hierarchical clustering methods. Non-hierarchical methods. Dendrogram. How to perform cluster analysis using computer package. Interpretation of computer output. Business applications such as market segmentation.
Multidimensional Scaling
Proximity measures. The classical (metric) solution. Nonmetric group methods: Shepard-Kruskal algorithm. Rotating the solution. Optimal properties and goodness of fit measures. How to perform multidimensional scaling using computer package. Interpretation of computer output. Business applications such as product positioning.