COURSE: Analysis of Panel Data

Course content

The purpose of the course is to provide an overview of statistical methods appropriate for the analysis of longitudinal data, or data collected on multiple units (individuals, states, dyads, countries) at more than one point in time. The lectures will focus on models for the analysis of “panel data,” which (by  convention) is used to describe data with relatively large number of units and relatively few time points. The course will begin by situating panel analysis  within alternative frameworks for causal inference. It will then provide an  overview of panel models from each of the three traditional approaches that dominate the field: the “econometric” tradition emphasizing unobserved heterogeneity; the “structural equation” tradition emphasizing models with reciprocal causality and measurement error, and the “multilevel modeling” tradition emphasizing models with longitudinal growth and random coefficients.

Education

Elective course for Security Risk Management

 

Bachelorlevel: 10 ECTS


Masterlevel: 7,5 ECTS

Learning outcome

Knowledge and understanding

• Taking advantage of the benefits that panel data provide the researcher in  making inferences about causal dynamics.

• Being sensitive to the specific problems and complexities that emerge when conducting panel analyses.

• Understand the relationship and distinctions between different methods for  analyzing panel data

Skills and abilities

• Model directly individual--‐level change and growth in dependent variables;

• Estimate models that control for the potential biases in causal inference  induced by unmeasured unit--‐specific effects or “nobserved heterogeneity;”

• Test alternative lag structures and models of reciprocal causality between variables;

• Estimate causal effects after controlling for the confounding effects of measurement error;

• Estimate models that specify and account for variation in individual--‐level intercepts, slopes and/or rates of change over time

Competencies

• Critically discuss caveats and advantages of different approaches and  techniques used for modelling and analysing panel data.

• Develop competence in the use of STATA and other statistical software  packages used in panel data analysis

The course will consist of lectures on the statistical material, in-­‐class exercises, written assignments, and student presentation of their final papers.

The main texts for the course will be:

Allison, Paul. 2009. Fixed Effects Regression Models. Thousand Oaks, Ca.: Sage.

Andreß, Golsch, and Schmidt. 2013. Applied Panel Data Analysis for Economic and Social Surveys. Berlin  Heidelberg: Springer--‐Verlag.

Finkel, Steven E. 1995. Causal Analysis with Panel Data. Thousand Oaks, Ca.: Sage.

Newson, Jason. 2015. Longitudinal Structural Equation Modeling. New York: Routledge Press.

Additional articles will be assigned and made available by 1 March 2017.

Familiarity with basic regression models for continuous and categorical dependent variables.

ECTS
7,5 ECTS
Type of assessment
Written examination
Written
Marking scale
7-point grading scale
Censorship form
External censorship
Criteria for exam assessment
  • Grade 12 is given for an outstanding performance: the student lives up to the course's goal description in an independent and convincing manner with no or few and minor shortcomings
  • Grade 7 is given for a good performance: the student is confidently able to live up to the goal description, albeit with several shortcomings
  • Grade 02 is given for an adequate performance: the minimum acceptable performance in which the student is only able to live up to the goal description in an insecure and incomplete manner

Single subject courses (day)

  • Category
  • Hours
  • Class Instruction
  • 28
  • English
  • 28