Econometrics: ECOM30002 and ECOM90002 Lecture 1 Maximum Likelihood: Basics Professor Vance L. Martin Semester 1 Professor Vance L. Martin () Econometrics: ECOM30002 and ECOM90002 Lecture 1 Maximum Likelihood: Basics Semester 1 1 / 48
Econometrics: ECOM30002 and ECOM90002Lecture 1 Maximum Likelihood: Basics
Professor Vance L. Martin
Semester 1
Professor Vance L. Martin () Econometrics: ECOM30002 and ECOM90002 Lecture 1 Maximum Likelihood: BasicsSemester 1 1 / 48
Lecture 1 Maximum Likelihood: Basics
Learning Objectives
1 Course Information2 Major Project3 The Linear Regression Model Reviewed4 Data Characteristics
Background Reading
1 �GROUP_PROJECT.PDF�, Handout on course website.
Professor Vance L. Martin () Econometrics: ECOM30002 and ECOM90002 Lecture 1 Maximum Likelihood: BasicsSemester 1 2 / 48
Lecture 1 Maximum Likelihood: BasicsCourse Information
Coordinator: Professor Vance L. Martin
Prerequisites:
One of Intermediate Macroeconomics or Intermediate Microeconomicsor Business Finance.One of Introductory Econometrics, Basic Econometrics, Statistics, oran H2A or better in Quantitative Methods 2.
Contact:
Lectures: Two 1-hour lecturesOne 1-hour tutorial per week beginning in the �rst weekTutor consultation times (times to be arranged - see course website)On-line tutor (available from the course website)Audio versions of the lectures are available on the web
Professor Vance L. Martin () Econometrics: ECOM30002 and ECOM90002 Lecture 1 Maximum Likelihood: BasicsSemester 1 3 / 48
Course Outline:
The course focuses on maximum likelihood methods to estimate and testeconometric models. Both analytical and numerical methods of estimationare discussed, while likelihood ratio, Wald and Lagrange multiplier testingframeworks are presented. Important special examples investigated areprobit and ordered probit regressions, Poisson regressions, linear andnonlinear regression models with heteroskedasticity and autocorrelation,ARCH models of volatility and simultaneous equations models. Allmathematical techniques including multivariate calculus and matrices, aredeveloped in the course.
Source: Melbourne Age, 3/3/2013
Professor Vance L. Martin () Econometrics: ECOM30002 and ECOM90002 Lecture 1 Maximum Likelihood: BasicsSemester 1 4 / 48
Main Reading:Lecture NotesLecture HandoutsArticles placed on the website
Background Reading:Hill, R.C., Gri¢ ths, W.E. and Lim, G.C. (2007), Principles ofEconometrics, 3rd Edition, Wiley.Stock, J.H. and Watson, M.W. (2007), Introduction to Econometrics,2nd Edition, Addison-Wesley Longman, Amsterdam.
Prescribed Text:None.Additional Reading:
Econometrics: Martin, V.L., Hurn, S. and Harris, D. (2013),Econometric Modelling with Time Series: Speci�cation, Estimation andTesting, Themes in Modern Econometrics, Cambridge University Press,New York.Calculus: Stewart, James (2008), Calculus, (6th ed.) ThomsonPublishers.Matrices: Anton, H. and Rorres, C. (2005), Elementary LinearAlgebra, Applications Version, (9th ed.) John Wiley and Sons.
Professor Vance L. Martin () Econometrics: ECOM30002 and ECOM90002 Lecture 1 Maximum Likelihood: BasicsSemester 1 5 / 48
CalculatorsStandard ordinary scienti�c calculators are allowed. Graphic andprogrammable calculators are not allowed in the exam.
Note that calculators will be needed in the �nal exam.
Software and Computers
EViews is available:
On campus in tutorial rooms and the Faculty computer lab.O¤ campus through a link on the course website.
Note that it will be necessary to be able to interpret EViews output inthe �nal exam.
Professor Vance L. Martin () Econometrics: ECOM30002 and ECOM90002 Lecture 1 Maximum Likelihood: BasicsSemester 1 6 / 48
Course Outline
Week Lectures Topic1 1,2 Maximum Likelihood: Basics2 3,4 Maximum Likelihood: General Case3 5,6 Maximum Likelihood: Numerical Methods4 7,8 Maximum Likelihood: Testing
5 9,10 Maximum Likelihood: Microeconometrics6 11,12 Maximum Likelihood: Heteroskedasticity7 13,14 Maximum Likelihood: Autocorrelation8 15,16 Maximum Likelihood: Speci�cation Analysis
9 17,18 Introduction to Simultaneous Equations10 19,20 Introduction to Simultaneous Equations11 21,22 Introduction to Asymptotic Distribution Theory12 23,24 Course Prizes, Exams and the Future
Professor Vance L. Martin () Econometrics: ECOM30002 and ECOM90002 Lecture 1 Maximum Likelihood: BasicsSemester 1 7 / 48
Weeks 1 to 4: Principles of Maximum Likelihood Estimation
Speci�cation
Estimation
Numerical Methods
Testing
Professor Vance L. Martin () Econometrics: ECOM30002 and ECOM90002 Lecture 1 Maximum Likelihood: BasicsSemester 1 8 / 48
Weeks 5 to 7: ApplicationsWho will win the Federal election?
Source: Melbourne Age, 18/2/2013, p.1
Professor Vance L. Martin () Econometrics: ECOM30002 and ECOM90002 Lecture 1 Maximum Likelihood: BasicsSemester 1 9 / 48
Source: Melbourne Age, 3/3/2013
Professor Vance L. Martin () Econometrics: ECOM30002 and ECOM90002 Lecture 1 Maximum Likelihood: BasicsSemester 1 10 / 48
What determines monetary policy?
Professor Vance L. Martin () Econometrics: ECOM30002 and ECOM90002 Lecture 1 Maximum Likelihood: BasicsSemester 1 11 / 48
Weeks 5 to 7: Applications continued
Modelling the Medal Count at the Olympic Games?
Predicting the medal count.
Professor Vance L. Martin () Econometrics: ECOM30002 and ECOM90002 Lecture 1 Maximum Likelihood: BasicsSemester 1 12 / 48
Weeks 5 to 7: Applications continued
The �Great Moderation�
Source: www.federalreserve.gov/BOARDDOCS/SPEECHES/2004
/20040220/default.htm
Professor Vance L. Martin () Econometrics: ECOM30002 and ECOM90002 Lecture 1 Maximum Likelihood: BasicsSemester 1 13 / 48
Weeks 5 to 7: Applications continued
The Food Crisis
Source: Melbourne Age, 15/4/2008, p.1
Professor Vance L. Martin () Econometrics: ECOM30002 and ECOM90002 Lecture 1 Maximum Likelihood: BasicsSemester 1 14 / 48
Weeks 5 to 7: Applications continuedClimate Change
Source: Melbourne Age, 21/4/2008, p.13
Which could have been avoided by doing THIS COURSE!Professor Vance L. Martin () Econometrics: ECOM30002 and ECOM90002 Lecture 1 Maximum Likelihood: BasicsSemester 1 15 / 48
Weeks 8 to 10: Extensions
Tru es
Source: Melbourne Age, Epicure, 10/2/2009, p.8
Professor Vance L. Martin () Econometrics: ECOM30002 and ECOM90002 Lecture 1 Maximum Likelihood: BasicsSemester 1 16 / 48
Weeks 11 to 12: Asymptotic Distribution Theory, Presentations andExam Preparation
We bring home the bacon!
Professor Vance L. Martin () Econometrics: ECOM30002 and ECOM90002 Lecture 1 Maximum Likelihood: BasicsSemester 1 17 / 48
Assessment:
Assessment Weight Due Where
Tutorial homework I 2.5% Week 3 Tutorial RoomTutorial homework II 2.5% Week 4 Tutorial RoomTutorial homework III 2.5% Week 5 Tutorial RoomTutorial homework IV 2.5% Week 6 Tutorial RoomTest (50 mins) 5% Week 7 Tutorial RoomMajor Project 20% Week 9 Submit onlineFinal Exam (2 hours) 65% End of Semester Exam Room
Professor Vance L. Martin () Econometrics: ECOM30002 and ECOM90002 Lecture 1 Maximum Likelihood: BasicsSemester 1 18 / 48
Lecture 1 Maximum Likelihood: BasicsMajor Project
Details
i Due Wednesday of Week 9 by 4.00pm (submit online).
ii Worth 20%
iii Groups sizes can be a minimum of one and a maximum of 6. Peopledo NOT have to be in the same tutorial.
iv AIM: Use the techniques of the course to write a research paper on atopic chosen by the group.
v You need to formulate your own hypotheses, collect the data,undertake empirical analysis and write up the results.
vi The expected length of the research report should be about 10 pages,with the maximum length not exceeding 15 pages, including allgraphs and tables.
vii Group decides on the weight allocated to each member of the group(stop free-riders). Any issues please see me!
Professor Vance L. Martin () Econometrics: ECOM30002 and ECOM90002 Lecture 1 Maximum Likelihood: BasicsSemester 1 19 / 48
Lecture 1 Maximum Likelihood: BasicsMajor Project
Timelinei Week 3: Form a group of at most 6 people and identify a topic.ii Week 4: Discuss the topic with your tutor in the tutorial(s) orconsultation times.
iii Week 5: Collect data.iv Week 6: Write-up the introduction.iv Week 7: Generate empirical results and write-up the results.v Week 8: Write-up the implications of the results and formulate theconclusions, together with an overall proofread of the project.
vi Week 9: Submission of project.vii Week 10: Research essays marked and compared across tutorials.viii Week 11: Projects are short-listed with names of projects given in the
Lecture and placed on the website.xi Week 12: In the Monday lecture of Week 12, Groups on the short-listpresent their project at the lecture. The Student Body chooses thewinner!
Professor Vance L. Martin () Econometrics: ECOM30002 and ECOM90002 Lecture 1 Maximum Likelihood: BasicsSemester 1 20 / 48
Lecture 1 Maximum Likelihood: BasicsMajor Project
Short-list from last year:
1 �Who�s Next? An Econometric Guide to the 2012 Election�2 �What Determines European Citizens�Attitudes Towards BiodiversityTowards Biodiversity Conservation?�
3 �Do Have What it Takes to Score A Second Date?�4 �An Analysis on Children�s Wellbeing in Rural India�5 �The Relationship Between the ASX200 and the Indicies of Its Top 5Trading Partners, the T-Distribution Approach�
Further details on the Major Project and other examples of potentialprojects, are given on the website in GROUP_PROJECT.PDF.
Professor Vance L. Martin () Econometrics: ECOM30002 and ECOM90002 Lecture 1 Maximum Likelihood: BasicsSemester 1 21 / 48
Lecture 1 Maximum Likelihood: BasicsMajor Project
The Prize:An all expenses paid dinner immediately after the last lecture on theWedneday of the course with
1 Your favourite econometrics lecturer.
AND
2 Invited eminent guests.
Past examples are
Professor Vance L. Martin () Econometrics: ECOM30002 and ECOM90002 Lecture 1 Maximum Likelihood: BasicsSemester 1 22 / 48
Sir Clive Granger
1 Nobel Prize winner in economics in 2003.2 Knighted by the queen.3 100 Welsh Heros list.
Professor Vance L. Martin () Econometrics: ECOM30002 and ECOM90002 Lecture 1 Maximum Likelihood: BasicsSemester 1 23 / 48
Source: http://www.100welshheroes.com/en/top100
Professor Vance L. Martin () Econometrics: ECOM30002 and ECOM90002 Lecture 1 Maximum Likelihood: BasicsSemester 1 24 / 48