Stat 543: Applied Multivariate Statistical Analysis

                                                  Spring 2012

Week

Date

Topics

Sections in the Book

Homework

Homework Due Date

Announcements

1

01/23/12

Syllabus, Tentative Schedule, Application of Multivariate Methods,

1.1, 1.2

2.1, 2.2 , 2.3, 2.4*, 2.5, 2.6*

01/30/12

  • Chapter 1--- Scripts
  • The starred problems are for Mathematical Sciences students ONLY.
  • Download the data sets from the text’s website

01/24/12

Data Organization,

 1.3

01/25/12

R-Script for describing data, Basics of Matrix Algebra

1.3, 1.4, 2.1, 2.2

 

 

01/27/12

Basics of Matrix Algebra,

2.2

 

2

01/30/12

Positive Definite Matrices, Spectral Decomposition, Square Root and Power Matrices

2.3, 2.4

2.11 (do not use the hint), 2.21, 2.25, 2.27, the two problems assigned in class*

02/10/12

 

01/31/12

Examples, Random vector and Matrices, mean vector, covariance and correlation matrices,

2.5, 2.6

 

 

 

02/01/12

Covariance and correlation matrices, Linear Combinations, Random Sampling

2.6, 3.1, 3.3

 

 

 

02/03/12

Generalized Variance, Multivariate Statistics as Matrix Operations, Singularity of S,

3.4, 3.2, 3.5

 

 

 

3

02/06/12

Statistics for Linear Combinations, The Geometry of the Sample,

3.6,

3.2, 3.7,3.15, 3.18, the two starred problems assigned in class

2/17/12

(1)   A2=A implies rank(A)=tr(A).

(2)    

02/07/12

The Multivariate Normal Distribution: Pdf, mgf, Linear Combinations, Marginal

4.2

 

 

 

02/08/12

The Multivariate Normal Distribution: Independence, Conditionals; Statistical Distance; Maximum Likelihood Estimation

4.2, 4.3

 

 

 

02/10/12

Sampling Distributions of Basic Multivariate Statistics, Large Sample Properties, Assessing Multivariate Normality

4.4, 4.5, 4.6

 

 

·         Chapter 4--- R-Scripts

4

02/13/12

Using R for Assign Normality, Chi-Squared Plot, Tests for Multivariate Normality

4.6

 4.29,  4.36,  4.37, 4.39

2/27/12

(1)   Likelihood function for Power transformation

(2)   One of the problems from the results for the Multivariate Normal Distribution

02/14/12

Mardia’s (1972, 1974), Royston’s (1983), Henze and Zirkler’s (1990),   Szekeley and Rizzo (2006) Tests for Multivariate Normality, Detecting Outliers, Box-Cox Transformation

4.7, 4.8

 

 

·         C. J. Mecklin and D. J. Mundfrom (2005). A Monte Carlo Comparison of the Type I and Type II Error Rates of Tests of Multivariate Normality, Journal of Statistical Computation and Simulation, 75, 93-107.

 

02/15/12

Test for the Mean Vector, Confidence Regions

5.1, 5.2, 5.3, 5.4

 

 

·         Chapter 5--- R-Scripts

02/17/12

Confidence Regions,  Simultaneous Confidence Intervals

5.4

 

 

 

5

02/20/12

President’s Day---No Class.

02/21/12

Large Sample Inference,  Missing Data Mechanisms

5.5, 5.7

 

 

 

02/22/12

Inference about the Mean Vector in the Presence of Missing Data

5.7

 

 

 

02/24/12

Missing Data, The Assumption of Independence, Multivariate Paired-Sample Test, Repeated Measures Analysis

5.7, 5.8, 6.1, 6.2

5.9, 5.15*, 5.16*,  5.18, 5.20, 5.23

03/02/12

·         Chapter 6--- R-Scripts

6

02/27/12

Repeated Measures Analysis, R-Example

6.2

 

 

 

02/28/12

R-Example, Inference with Two Independent Samples

6.2, 6.3

 

 

 

02/29/12

Inference with Two Independent Samples

6.3

 

 

 

03/02/12

Behrens-Fisher Problem, Large Sample Inference

6.3

·         6.16, 6.19, 6.20, 6.21. For Statistics Students (1) Derive LRT for testing H_0: C mu=0. (2) T^2 is invariant to the choice of contrast matrix.

03/09/12

 

7

03/05/12

One-Way MANOVA,

6.4

 

 

03/06/12

One-Way MANOVA, Simultaneous Inference

6.4, 6.5

 

 

 

03/07/12

Simultaneous Inference, Box’s M-Test for equality of Covariance Matrices

6.5, 6.6

 

 

 

03/09/12

Two-way MANOVA

6.7

6.25, 6.27, 6.31, 6.33

 

·         For Statistics Students: (1) Derive LRT for testing H0:mu_1-mu_2=delta_0 when samples come from two normal populations with equal covariance matrix  (2) Show that Wilks’ Lambda Statistis for one way MANOVA depends on the data only through the eigenvalues of BW^{-1}.

8

03/12/12

Two-way MANOVA, Profile Analysis

6.7, 6.8

 

 

·         Srivastava, M.S. (1987). Profile Analysis of Several groups. Comm. in Stat. Theory & Methods. 16, 909-926.

03/13/12

Profile Analysis

 6.8

 

 

 

03/14/12

Profile Analysis, Growth Curve Analysis

6.9

 

 

 

03/16/12

Growth Curve Analysis

6.9

 

 

 

9

03/19/12

Multivaraite Multiple Regression

7.7

6.39, Handout, 7.25, 7.26

03/27/12

·         Chapter 7--- R-Scripts

03/20/12

Multivariate Multiple Regression

7.8

 

 

 

03/21/12

Multivariate Multiple Regression

7.9

 

 

 

03/23/12

Population Principal Components Analysis

8.1, 8.2

 

 

 

10

03/26/12

Population PCA,  PCA under Normality, Standardizing the Variables

8.2

8.13  (a), (b) and (c), 8.22, 8.26, 8.27

04/10/12

For Statistics Students:

(1)    Find the eigenvalues for the equi-correlation covariance structure.

(2)    Do problem 8.9.

03/27/12

PCA for Covariance with Special Structure, Sample PCA

8.2

 

 

·         Chapter 8--- R-Scripts

03/28/12

Sample PCA, Examples--Turtle Data, WAIS Data, 1984 Olympic  Women’s Athletics data

8.3

 

 

 

03/30/12

Graphing the PCs, Large Sample Inference

8.4, 8.5

 

 

 

 

11

04/09/12

Large Sample Inference, Canonical Variates and Canonical Correlations

8.5, 10.1

10.7*, 10.9, 10.12, 10.16, 10.18, One starred problem was assigned in class

 

 

·         Chapter 10--- R-Scripts

04/10/12

Standardization, Interpreting the Canonical Variates, Sample Canonical Variates and Correlations, Examples

10.2

 

 

 

04/11/12

Examples, Matrices of Errors of Approximation

10.2, 10.3

 

 

 

04/13/12

Proportion of Explained Sample Variance, Large Sample Inference

10.3, 10.4, 10.5

 

 

 

12

04/16/12

Separation and Classification for Two Populations, Minimum ECM and TPM classifiers

11.1, 11.2

 

 

·         Chapter 11--- R-Scripts

04/17/12

Classification for Two Multivariate Normal Populations with Equal and Unequal Covariances, Fishers Approach

11.3

 

 

 

04/18/12

Fishers Approach , Evaluating Classification Functions

11.3, 11.4

 

 

 

04/20/12

Evaluating Classification Functions, Example

11.4

 

 

 

13

04/23/12

Classification with Several Populations

11.5

 

 

 

04/24/12

Classifications with Several Populations, Example

11.5

 

 

 

04/25/12

Example Continued, Fisher’s Approach

11.5, 11.6

 

 

 

04/27/12

Fisher’s Approach, Example, Dimensionality, Large Sample Inference

11.6, 8.6 from Rencher(2002)

11.7*, 11.24, 11.30, 11.33, 11.35

 

 

·         Section 8.6 from A. C. Rencher, Methods of Multivariate Analysis, Wiley 2002.

14

04/30/12

 

 

 

 

 

05/01/12

 

 

 

 

 

05/02/12

 

 

 

 

 

05/04/12

 

 

 

 

 

15

05/11/12