To complete this unit you need to
- Watch the video lectures
- Read the materials for the written assignment 4 (optional)
- Complete the unit 5 discussion forum task
- Return written assignment 4 (optional)
- View the model answer for written assignment 4 and instructors comment's to written assignment 4 (optional)
- Participate in the seminar (optional)
The unit introduced the generalized linear model, which is an extension to linear regression covering most commonly used single dependent variable models as special cases (e.g. logistic regression, poisson regression, tobit regression, etc.). Maximum likelihood estimation is introduced. After this unit you should:
- Understand that the reason for using GLM models is that you want to model non-linear relationships. While GLM models are often used when the dependent variable is non-continuous (binary, count, categorical), the distribution of the dependent variable is not a reason to use GLM per se, but non-linear models often make sense in this case.
- Understand the interpretation of the three most common GLM curves: the line (normal regression), the exponential curve (log link), the S-curves (logistic and probit) and how the effects of other variables are interpreted when these curves are used.
- Understand why plotting is essential when interpreting non-linear models.
- Have a basic understanding of the most common combinations of link and distribution and when they would be used: logistic and probit regression, ordinal and multinomial regression, poisson and negative binomial regression. A through understanding of these techniques and their special interpretations or diagnostics is not required.
- (Only students that want to use statistical analysis techniques in their own research:) Understand the principle of maximum likelihood estimation and have a basic idea of some of the computational issues that may arise - this is important because you will encounter these issues in your own research.
This units includes a lot of technical details of many commonly used modeling approaches. You do not need to understand all these details unless you plan to use these techniques yourself. Particularly the videos under "Topic 3: S-curve and exponential models for different kinds of dependent variables" are material that you do not need to understand in detail. These videos are included to give you ideas on what you could use in your own research, to explain how and why plotting is essential for interpreting these models (and to point out that most published papers fail to do so), and to provide a reference point where to start if you decide that you want to know more about a specific model.