MET522 Multivariate Data Analysis

• Topics

  The course is an applied course in multivariate analysis, with focus on structural equation modeling. Teaching idea is "Learning by doing". I.e., the students will work with real data and software (R/lavaan) to motivate for statistical theory, interpretation and learning to be aware of challenges when statistical (mathematical) theories meet real data.

  • OLS Regression Analysis including ANOVA and ANCOVA,
  • Classification: Logistic Regression, Linear and Quadratic Discriminant Analysis,
  • Instrumental Variables and 2SLS,
  • The Multivariate Linear Model,
  • Exploratory - and Confirmatory Factor Analysis,
  • Structural Equation Models,
  • Analysis of Longitudinal Data,
  • Multiple Group analysis.

• Learning outcome

  After completion of the course, the students will be able to:

  Knowledge:

  *Apply and understand some classical methods in multivariate statistics

  *Develop their own statistical "tool box" based on existing knowledge in the field.

  Skills:

  *Apply modern statistical softwares and apply these on research projects.

  *Developed an understanding of SEM by relating it to participants' previous knowledge of multiple linear regression, and then by expanding it to allow for correlated and causally related latent constructs.

  *Apply methods such as path analysis among measured variables, confirmatory factor models, structural models involving latent causality, and multi-group models and a preview of more advanced topics. Examples from a variety of disciplines will be accompanied.

  General competence:

  *Discuss and raise questions on methods in multivariate statistics.

• Teaching

  Lectures and software demos.The course is divided into two seminars - each lasting 3 or 4 days. The couse will be offered digitally if it cannot be held on campus.

• Restricted access

  PhD candidates from NHH as well as PhD candidates from other institutions can take part in the course.

  Motivated master's students at NHH may be admitted after application, but are subject to the approval from the course responsible on a case by case basis. Individuals outside academia may be admitted after application, but are subject to the approval from the course responsible and the Vice Rector for Reserach on a case by case basis.

  There is no cap on the number of students.

  NHH Research Scholars should register for the course as recommended.

  External Research Scholars: We kindly request you send an application to attend the course https://www.nhh.no/en/study-programmes/phd-programme-at-nhh/phd-courses/become-a-visiting-student-at-a-phd-course-at-nhh/See link to information here within 3 weeks before the start of the course.

• Required prerequisites

  Basic course in statistics

• Credit reduction due to overlap

  None

• Compulsory Activity

  Mandatory attendance.

• Assessment

  Individual written term paper.

• Grading Scale

  A-F

• Computer tools

  R-lavaan/LISREL/Mplus

• Literature

  Jöreskog, Olsson, and Wallentin (2016): Multivariate Analysis with LISREL. 1st ed. Springer International Publishing

  Recommended reading:

  James, Witten, Hastie and Tibshirani (2017): An Introduction to Statistical Learning with Applications I R. Springer Texts in Statistics

  Handouts in multivariate statistics (under preparation)