Vær opmærksom på at dette website indeholder et arkiv med historiske data. Det aktuelle kursuskatalog findes på kursuskatalog.au.dk

AU kursuskatalog arkiv

[Forside] [Hovedområder] [Perioder] [Udannelser] [Alle kurser på en side]

Regression Analysis ( forår 2008 - 5 ECTS )

Rammer for udbud

  • Uddannelsessprog: (se under Undervisnings- og arbejdsform)
  • Niveau: Compulsory 4th semester course in the BSc (Econonomics and Management) program
  • Semester/kvarter: Every Spring
  • Timer per uge: 4 lectures + 2 exercise sessions for 7 weeks  
  • Deltagerbegrænsning: None
  • Undervisningssted: Århus
  • Hovedområde: Det Samfundsvidenskabelige Fakultet
  • Udbud ID: 10907

Formål

The objectives of the course are that the participants can:

·         formulate the multiple linear regression model and its underlying assumptions in matrix form

·         estimate and interpret the parameters of multiple linear regressions

·         describe the statistical properties of the estimated parameters

·         apply hypothesis testing concerning the parameters of multiple regression models

·         apply the estimated regression model to compute forecasts and to interpret the precision of these forecasts

·         describe the consequences of multicollinearity, omitted variables, functional form misspecification, heteroskedasticity and autocorrelation in multiple regression models

·         evaluate the adequacy of the estimated regression models by performing test for omitted variables, functional form misspecification as well as neglected heteroskedasticity and autocorrelation

·         apply efficient estimation in regression models with heteroskedastic innovations

·         interpret and evaluate autocorrelation and heteroskedastic consistent statistics

·         apply statistical inference in dynamic multiple regressions.

Indhold

The course introduces the students to regression analysis and teaches them to apply basic statistical techniques to estimate key parameters in economic models for describing economic and operational relations. The course also covers use of various databases and statistical software, including PcGive and GiveWin for analysis of macroeconomic issues

 

COURSE SUBJECT AREAS

  • Regression analysis with several explanatory variables.
  • Model estimation and specification.
  • Hypothesis testing in multiple regressions.
  • Partial regression.
  • Functional form, dummy variables and parameter stability.
  • Heteroskedasticity.
  • Autocorrelation and dynamic regression models.

 

 

Faglige forudsætninger

None

Underviser

Christian M. Dahl

Undervisnings- og arbejdsform

The lectures will cover various theoretical topics as well as empirical illustrations. The software package OxMetrics/PcGive will be used extensively in the lectures throughout the course.

Litteratur

   

  • Heij, C., P. de Boer, P.H. Franses, T. Kloek & H.K. van Dijk (2004). Econometric Methods with Applications in Business and Economics. Oxford University Press , Ch. 3, pp. 117-186, Ch. 5, pp. 273-378, 386-388. . The total number of pages of required reading in Heij et al equals about 176 pages.
  • Various notes, approximately 50 pages.

 

The total number of pages of required reading for the course equals approximately 230.

 

 

Bedømmelse

A 4-hour written exam jointly with Surveys and Qualitative Data. Students who are not taking Surveys and Qualitative Data will only take a 2 hour-written exam in Regression Analysis. An approved obligatory assignment is a requirement to take the exam.

The obligatory assignment will be handed out and solved during the last week of the course. The obligatory assignment can be solved in groups of up to 4 persons.

EXAMINATION AIDS ALLOWED: All except any means of electronic communication including PCs