Course Information

Course Information

Course Title	Code	Semester	L+U Hour	Credits	ECTS
GENERALIZED LINEAR MODELS (PRACTICE)	23601048		0 + 2	1.0	3.0

Prerequisites

None

Language of Instruction	Turkish
Course Level	Graduate Degree
Course Type	Compulsory
Mode of delivery	Lecture,Q&A,Discussion,Practise
Course Coordinator
Instructors	BEYZA DOĞANAY ERDOĞAN
Assistants
Goals	To educate experts having the knowledge, skill and attitude to perform generalized linear models
Course Content	The Analysis of Variance Models, Analysis of Contingency Tables, Two-Way Models, Three-Way Models, Introduction to GLM, Continuous Response Models, Binomial Response Models, Ordered Response Models, Nominal Response Models, Count Response Models, Modelling Contingency Tables
Learning Outcomes	1) Knows the logic of statistical modelling. 2) Describes the uses of Generalized Linear Models. 3) Tests the assumptions of models. 4) Models the relationship between the dependent and independent variables. 5) Chooses the most appropriate model according to the type of dependent variable. 6) Interprets the model parameter estimates. 7) Interprets the significances of the model parameter estimates.

Weekly Topics (Content)

Week	Topics	Teaching and Learning Methods and Techniques	Study Materials
1. Week	Introduction, Data Types, Sample Data Sets	Lecture	Presentation (Including Preparation Time)
2. Week	Analysis of Variance Models (Generalized Linear Models) ANOVA, MANOVA, ANCOVA-MANCOVA, The Regression Approach to the ANOVA Models	Lecture	Presentation (Including Preparation Time)
3. Week	Analysis of Crosstabs; Two-way Crosstabs, Testing Statistical Independence, Odds Ratio and Relative Risk, Binomial, Multinomial and Poisson Distributions	Lecture	Presentation (Including Preparation Time)
4. Week	Analysis of Crosstabs; Three-way Crosstabs, Conditional and Marginal Odds Ratios, Conditional Independence - Marginal Independence, Simpson's Paradox	Lecture	Presentation (Including Preparation Time)
5. Week	Introduction to Generalized Linear Models; Model Components (Random Component, Linear Systematic Component, Link Function), Assumptions, Exponential Family	Lecture	Presentation (Including Preparation Time)
6. Week	Parameter Estimation Algorithms and Goodness of Fit Measures; Newton - Raphson algorithm, Fisher's Scoring algorithm, Variance Estimation Methods, Goodness of Fit Measures	Lecture	Presentation (Including Preparation Time)
7. Week	Continuous Response Models; The Gaussian Family, Linear Regression, Link Function, Statistical Inference and Model Selection Criterias, Example	Lecture	Presentation (Including Preparation Time)
8. Week	Binomial Response Models; Logistic Regression, Conditional Logistic Regression, Exact Logistic Regression	Lecture	Presentation (Including Preparation Time)
9. Week	Binomial Response Models; Probit Regression, The Log-log and Complementary Log-log Models	Lecture	Presentation (Including Preparation Time)
10. Week	Binomial Response Models; Overdispersion problem, Statistical Inference and Model Selection Criterias	Lecture	Presentation (Including Preparation Time)
11. Week	Binomial Response Models, Example	Lecture	Presentation (Including Preparation Time)
12. Week	Ordered Response Models; Ordinal Logistic Regression, Ordinal Probit Regression	Lecture	Presentation (Including Preparation Time)
13. Week	Ordinal Response Models; Statistical Inference and Model Selection Criterias, Example	Lecture	Presentation (Including Preparation Time)
14. Week	Nominal Response Models; Multinomial Logistic Regression, Statistical Inference and Model Selection Crtiterias, Example	Lecture	Presentation (Including Preparation Time)
15. Week	Count Response Models; Poisson Regression, Negative Binomial Regression, Statistical Inference and Model Selection Criterias, Example	Lecture	Presentation (Including Preparation Time)
16. Week	Modelling crosstabs, Log-linear Models, Statistical Inference and Model Selection Criterias	Lecture	Presentation (Including Preparation Time)

Sources Used in This Course

Recommended Sources

Dobson A.J.: An Introduction to Generalized Linear Models, 2002, Chapman and Hall, Florida.

McCullagh, P.; Nelder, J.A.: Generalized Linear Models, 1989, Chapman and Hall, Florida.

Relations with Education Attainment Program Course Competencies

Program Requirements	Contribution Level	DK1	DK2	DK3	DK4	DK5	DK6	DK7
PY1	5	0	0	0	0	0	0	0
PY2	5	0	0	0	0	0	0	0
PY3	5	0	0	0	0	0	0	0
PY4	5	0	0	0	0	0	0	0

*DK = Course's Contrubution.

	0	1	2	3	4	5
Level of contribution	None	Very Low	Low	Fair	High	Very High

.

ECTS credits and course workload

Event	Quantity	Duration (Hour)	Total Workload (Hour)
Course Duration (Total weeks*Hours per week)	14	2	28
Work Hour outside Classroom (Preparation, strengthening)	14	1	14
Homework	2	4	8
Presentation (Including Preparation Time)	2	6	12
Report (Including Preparation and presentation Time)	2	10	20
Final Exam	1	1	1
Time to prepare for Final Exam	1	8	8
Total Workload
Total Workload / 30 (s)
ECTS Credit of the Course			30

Quick Access

Biostatistics (Ph.D.)
Doktora

Course Information

Course Information
Weekly Topics (Content)
Sources Used in This Course
Relations with Education Attainment Program Course Competencies
ECTS credits and course workload