Syllabus
Mathematical Statistics
1. Name of the Course: Mathematical Statistics
• Lecturer: Péter Elek
• Department: Department of Economics
• Semester and year: Fall term, 2014/2015
• Course level: MA 1st year core (compulsory) course
• Number of credits: 3 CEU credits (6 ECTS credits)
• Pre-requisites: Pre-session Mathematics
• Course e-learning site: [Link]
• Office hours: T. B. A.
2. Course Description: The course discusses the basic statistical theory that is
frequently used in econometric analysis. Students passing the exam should
be familiar with the basic statistical concepts used intensively in economet-
rics, so that the later course (Econometrics I.) could focus on econometric
issues instead of the technicalities.
3. Learning Outcomes: A good understanding of the following topics:
• Essentials in probability theory
• Samples and sampling distributions
• Estimators
• Finite sample properties of estimators
• Asymptotic properties of estimators
• General methods to prepare estimators
• Interval estimation
• Hypothesis testing
4. Course Requirements:
• Problem sets during the term with a total weight of 30%. Some prob-
lem sets should be solved individually while others will be assigned to
the study groups, where one solution from each group is expected to
be submitted.
1
• Final exam for a total weight of 70%
• The final course grade will be assigned on a curve.
5. Assumed Background:
• Basic calculus (at the level of pre-session maths)
• Basic linear algebra (at the level of pre-session maths)
6. Textbooks:
• Jeffrey Wooldridge: Introductory Econometrics, A Modern Approach,
4th edition, 2009 (later denoted by W).
– The Appendix of this textbook contains an intuitive overview of
the basic statistical concepts that are used later in Econometrics I.
• Takeshi Amemiya: Introduction to Statistics and Econometrics, 1994
(later denoted by A).
– A good graduate level introduction to statistics. Not all details
are needed in the course (e.g. proofs are not required and some
examples are merely illustrative).
• Charles Henry Brase and Corrinne Pellillo Brase: Understandable Sta-
tistics, 9th edition, 2009 (later denoted by BB).
– Recommended for those who need a bit more help for understand-
ing the basic concepts.
Course schedule
WEEK 1: Essentials in probability theory
• Random variables, probability distributions
• Measures of central tendency and variability; higher order moments
• Other descriptive measures of distributions
• Joint distributions, conditional distributions, definitions of independence
• Conditional expectation, covariance, correlation
• Frequently used distributions and their properties
• The normal distribution and its properties
2
• Chi-squared, t- and F-distributions and their properties
• Material:
– W Appendix B
– A 1-5 (except for the proofs, the material around Theorem 3.6.3, sec-
tion 3.7, definition 5.3.1, section 5.4)
WEEK 2: Samples, estimators and their finite sample properties
• Populations and samples
• Different types of samples
• Sample statistics and sampling distributions
• Properties of sample means and variances
• Sampling from the normal distribution
• Parameters and estimators
• Unbiasedness
• Efficiency
• Comparison of estimators: Mean Squared Error (MSE)
• Best unbiased estimators
• Best linear unbiased estimators (BLUE)
• Examples
• Material:
– W Appendix C.1-C.2
– A 1, 7.1-7.2 (except for sections 7.1.3, 7.2.2, 7.2.4, 7.2.6)
3
WEEK 3: Asymptotic Properties of Estimators
• Asymptotic unbiasedness of an estimator
• Probability limit (convergence in probability) and its properties
• Continuous mapping theorem
• Weak law of large numbers
• Consistency of an estimator
• Convergence in distribution and its properties
• Central limit theorem
• Asymptotic normality of an estimator
• Asymptotic variance and asymptotic efficiency of an estimator
• Examples
• Material:
– W Appendix C.3
– A 6 (except for definition 6.1.3, Theorem 6.1.1, Theorem 6.2.3, Example
6.4.2) and section 7.2.6
WEEK 4: General Approaches to Parameter Estimation
• Method of moments (MM-) estimators
• Maximum likelihood (ML-) estimators and its properties
• Least squares (LS-) estimators
• Different estimators of the parameters of frequently used distributions
• Examples
• Material:
– W Appendix C.4
– A 7.1.1, 7.3, 7.4.4 (and the remaining parts of 7.4 for the supplemen-
tary material)
4
WEEK 5: Interval Estimation
• Concept of interval estimation
• Confidence intervals for means and variances from a normal distribution
• Determination of the sample size
• Asymptotic confidence intervals for non-normal (e.g. binomial) distribu-
tions
• Examples
• Material:
– W Appendix C.5
– A 8.1-8.2
WEEK 6: Hypothesis Testing
• Concept of hypothesis testing
• Null- and alternative hypothesis
• Type-I and type-II errors
• Significance levels and powers of the tests
• p-values
• Tests for the expected value and variance of random variables
• Relationship between confidence intervals and hypothesis testing
• Practical versus statistical significance
• Examples
• Material:
– W Appendix C.6
– A 9.1, 9.2, 9.6 (except for Theorem 9.2.1) and the beginning of 9.4