|
|||
|
|||
TEXT
OBJECTIVE
The course will study methods for the interpretation and analysis of multivariate and multi factor data. These techniques comprise the subject areas of regression and multivariate analysis. This will be a hands-on applied course, where we will learn statistical analysis by exposure to real data. Our focus will not be on mathematical derivations and proofs of theorems. The course is not however a cookbook course. We will learn not just to do things but why we do what we do. We will see the underpinnings of statistical theory for sound analysis.
The datasets analyzed will almost always be large so as to need the computer. In this course we will use Minitab. This package is available in the computer labs, and a Student version is available for purchase at the bookstore. I recommend that you purchase the package as almost all of the course work will involve analyzing data.
The course grade will be based on homework and projects only. Assignments must be typed, handwritten assignments will not be accepted. The final grade for the course will be on the homework and the assigned projects; there will be no opportunities for makeup or extra credit work. There will be three major projects in the course.
For the projects you will be responsible for collecting your own data. Do not use data from textbooks; get your data from original data sources. All assignments have to be handed in on time. Work-related travel will not be accepted as an excuse for lateness; it is your responsibility to get the assignment to me on time. Do not wait to do your assignment at the last moment, as you might find access to the School computing facilities at the last minute difficult. The lack of access will not be accepted as an excuse for lateness.
A good careful analysis of data requires time. This course will require a generous amount of time. More time you put into this course more you will get out of it. The skills learnt in this course will be of value in long years to come.
PREREQUISITE
Introductory statistics core course, and familiarity with the use of a statistical package.
SYLLABUS
| 1 | Basic concepts in simple regression | Chapter 2 | |
| 2 | Multiple regression | Chapter 3 | |
| 3 | Detection of model violations: Regression Diagnostics | Chapter 4 | |
| 4 | Qualitative variables as Predictors | Chapter 5 | |
| 5 | Checking model assumptions and correcting for violations | ||
| a. transforming variable | Chapter 6 | ||
| b. nonconstant variance | Chapter 7 | ||
| c. autocorrelation | Chapter 8 | ||
| d. multicollinearity | Chapter 9 | ||
| 6 | Choosing variables - model selection | Chapter 11 | |
| 7 | Modelling group membership: logistic regression, discriminant analysis | Chapter 12 | |
| 8 | Principal components analysis | Chapter 9 (Appendix) | |
| 9 | Factor Analysis | Chapter |
