Scribd
Upload
40 views3 pages

311 LO Part II

Uploaded by

u20417242
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
40 views3 pages

311 LO Part II

Uploaded by

u20417242
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

WST 311

Multivariate analysis
Lecture Objectives – Part 2
(2023)
Note that these lecture objectives act as a guide1 for you - it is not exhaustive. You need to add and amend to these
objectives as discussed in lectures. Please also make sure to consult the study guide for the broadly defined outcomes
per study theme.

4 Section 4: Introduction to Simple Linear Regression (SLR)


4.1 LECTURE 1, 2 and practical
OBJECTIVES
At the end of this lecture the student will be able to:
1) Defining the model and model assumptions of the model
a. Response variable, explanatory variable and residual term (error term)
2) Estimation of model parameters using the least-square principle
3) Construction of residual plots.
4) Using residual plots to check reasonability of the model assumptions
a. Constant variance and normal distribution

5) Sampling distribution of the estimator of the slope coefficient


a. Confidence interval for slope coefficient and Hypothesis testing
6) Coefficient of determination
7) Using SAS, Proc Reg and Proc GLM to fit the model to data
a. Writing code to fit the model
b. Interpretation of output

5 Section 5: Multiple Linear Regression (MLR)


5.1 LECTURE 3, 4 and practical
OBJECTIVES
At the end of this lecture the student will be able to:
1) Defining the MLR model and the assumptions for the model
a. Model in matrix notation, design matrix, parameter vector, etc.
2) Estimation of model parameters
a. Least square principle (OLS – ordinary least squares)
b. Maximum Likelihood principle (MLE)
3) Derive properties of parameter estimators and the sampling distributions of the estimators
a. Confidence interval for the parameters
b. Testing of hypothesis about parameters
4) Calculation and interpretation of the Coefficient of determination
5) Using SAS - Proc Reg, Proc GLM and Proc IML to fit a model to data
a. Interpretation of SAS software output.

6 Section 6: Multiple Linear Regression (MLR)


6.1 LECTURE 5, 6 and practical
OBJECTIVES
At the end of this lecture the student will be able to:
1) To do hypothesis testing of contrasts for
a. general linear hypothesis, single tests or simultaneous tests
b. linear combinations of regression parameters
2) Derive the null distribution of the test statistic for a general linear hypothesis.
3) Test of the overall regression, all coefficient are zero vs at least one is different from zero.
4) Using SAS, Proc GLM to execute the tests, setting up of Contrast statements
a. Interpretation of SAS software output

7 Section 7: Analysis of Variance, ANOVA


7.1 LECTURE 6, 7 and practical
OBJECTIVES
At the end of this lecture the student will be able to:
1) Perform Analysis of variance,
a. for the comparison of population averages
b. identify the ANOVA model as a special case of the MLR model
2) Calculate dummy variables
a. effect and reference constraint
3) Setting up of design matrix
a. full and not full rank using dummy variables
4) Testing ANOVA hypotheses
5) Using SAS, Proc GLM to
a. fit ANOVA models
b. perform hypothesis tests
• overall significance of the model
• pairwise comparisons
c. using the Class statement and/or dummy variables
d. Interpretation of SAS software output

8 Section 8: Analysis of Co Variance, ANCOVA


8.1 LECTURE 8, 9 and practical
OBJECTIVES
At the end of this lecture the student will be able to:
1) Analysis of co variance,
a. comparison of population averages with variance reduction using co variates.
2) ANCOVA model as a special case of a MLR model.
3) Setting up of design matrix
a. full and not full rank using dummy variables and including co variates.
4) Testing hypotheses
a. overall significance of the model
b. pairwise comparisons
5) Using SAS, Proc GLM to
a. fit ANCOVA models
b. perform hypothesis tests
i. overall significance of the model
ii. pairwise comparisons
iii. using the Class statement and/or dummy variables including co variates
c. Interpretation of SAS software output
9 Section 9 and 10: Generalised linear model, GLM
9.1 LECTURE 10, 11, 12 and practical
OBJECTIVES
At the end of this lecture the student will be able to:
1) Identify the Exponential family of distributions
a. Write distribution in the general exponential family
i. Special cases: Normal, Poisson, Bernoulli, etc
2) Defining the GLM
a. Random component
b. Systematic component
c. Link function
3) Use Deviance as a model fit measure
4) Fit GLM models
a. Poisson regression
b. logistic regression
5) Using SAS, Proc Genmod and Proc Logistic to fit GLM models
a. Poisson regression
b. logistic regression
c. Interpretation of SAS software output

You might also like