Description of Individual Course Units
Course Unit CodeCourse Unit TitleType of Course UnitYear of StudySemesterNumber of ECTS Credits
İST410ROBUST STATISTICSElective485
Level of Course Unit
First Cycle
Objectives of the Course
To make students learn the robust statistics concept and to give the students the opportunity to investigate the robust statistical methods in the literature and to compute and compare them by the help of the Matlab software.
Name of Lecturer(s)
Assoc. Prof. Dr. Hakan Savaş SAZAK
Learning Outcomes
1To Know the Basic Statistical Concepts
2To Know the Theoretical Statistical Concepts
3To Know the Statistical Estimation Methods
4To be able to derive estimators with various methods under different distributions
5To Know the Robust Statistics Concepts
6To Know Various Robust Statistics Methods
7To Know the Properties of Robust Estimators and Statistics
8To be able to implement the simulation of various statistical methods on computer
9To be able to compare the statistical methods under interest with respect to their properties and simulation results and make comments on them
10To be able to conduct the applications of Robust Statistical Methods
Mode of Delivery
Face to Face
Prerequisites and co-requisities
Recommended Optional Programme Components
Basic Usage Knowledge on the Matlab Software
Course Contents
• Basic Concepts: Statistic, estimator, unbiasedness etc. • Frequently used distributions and their properties • Properties of Estimators: Efficiency, unbiasedness, consistency etc. • The likelihood concept and the Fisher information matrix • Basic Estimation Techniques: Least Squares Estimators, Moment Estimators, Maximum Likelihood Estimators • Properties of Robust Statistical Methods • The concept of Order Statistics and their properties • Robust Estimation Methods: Weighted Least Squares, BLUE, L Estimators, Median, M-Estimators, Modified Maximum Likelihood Estimators • Measures of Robustness: Breakdown Bound and Influence Function • The simulation and comparison of Robust Estimators • Robust Regression
Weekly Detailed Course Contents
WeekTheoreticalPracticeLaboratory
0Basic Concepts: Statistic, estimator, unbiasedness etc.
1Frequently used distributions and their properties
2Properties of Estimators: Efficiency, unbiasedness, consistency etc.
3The likelihood concept and the Fisher information matrix
4Basic Estimation Techniques: Least Squares Estimators, Moment Estimators, Maximum Likelihood Estimators
5Properties of Robust Statistical Methods
6The concept of Order Statistics and their properties
7Midterm Exam
8Robust Estimation Methods: Weighted Least Squares, BLUE, L Estimators, Median, M-Estimators, Modified Maximum Likelihood Estimators
9Measures of Robustness: Breakdown Bound and Influence Function
10Dixon’s outlier model, mixed model, contamination model
11Obtaining various properties of estimators under different models
12Introduction of Matlab software and creating specialized programs by using Matlab
13The simulation and comparison of Robust estimators under different distributions and models
14Robust Regression
15Final Exam
Recommended or Required Reading
Hoaglin, D.C., Moesteller, F. & Tukey, J.W., “Understanding Robust and Exploratory Data Analysis”, 1983, John Wiley & Sons Inc. Martinez, W.L. & Martinez, A.R., “Computational Statistics Handbook with MATLAB”, 2002, Chapman & Hall/CRC
Planned Learning Activities and Teaching Methods
Activities are given in detail in the section of "Assessment Methods and Criteria" and "Workload Calculation"
Assessment Methods and Criteria
Term (or Year) Learning ActivitiesQuantityWeight
Midterm Examination1100
SUM100
End Of Term (or Year) Learning ActivitiesQuantityWeight
Final Sınavı1100
SUM100
Term (or Year) Learning Activities40
End Of Term (or Year) Learning Activities60
SUM100
Language of Instruction
English
Work Placement(s)
Workload Calculation
ActivitiesNumberTime (hours)Total Work Load (hours)
Midterm Examination122
Final Examination122
Attending Lectures14342
Individual Study for Mid term Examination14141
Individual Study for Final Examination16363
TOTAL WORKLOAD (hours)150
Contribution of Learning Outcomes to Programme Outcomes
PO
1
PO
2
PO
3
PO
4
PO
5
PO
6
PO
7
PO
8
PO
9
PO
10
PO
11
PO
12
PO
13
PO
14
PO
15
PO
16
PO
17
PO
18
PO
19
PO
20
PO
21
PO
22
PO
23
PO
24
LO13                       
LO22                       
LO32                       
LO42                       
LO5                        
LO6                        
LO72                       
LO82                       
LO9                        
LO10                        
* Contribution Level : 1 Very low 2 Low 3 Medium 4 High 5 Very High
 
Ege University, Bornova - İzmir / TURKEY • Phone: +90 232 311 10 10 • e-mail: intrec@mail.ege.edu.tr