Quantile Regression

Institution: see Organisers & Acknowledgements

Programme of study: International Research Workshop

Lecturer: PD Dr. Elke Holst (DIW Berlin & University of Flensburg), Andrea Schäfer, SOCIUM/Universität Bremen)

Date: Monday, 11/09/17 – Wednesday, 13/09/17 (09.00 – 12.30 h)

Max. number of participants: 20

Credit Points: 5 CP for participating in the whole IRWS

Language of instruction: English

Contents:

Quantile regression is a statistical analysis able to detect more effects than classical procedures, it does not restrict attention to the conditional mean and therefore it permits to approximate the whole distribution of a response variable. This course will introduce the basics of quantile regression methods and briefly discuss some recent developments, highlighting among others the differences between conditional and unconditional regression frameworks.

The course will further illustrate their application – discussing practical aspects of computation and inference – with the aim of providing practical guidance on the use of the methods in realistic applications. The application refers to the German wage structure using the German Socio-Economic Panel (SOEP). Thus, a substantial part of the course will cover an overview of the SOEP data structure and the research designs facilitated by longitudinal household studies that go beyond conventional surveys (household analysis, intergenerational analysis, life course research, etc.).

By the end of the course participants will:

  • Understand the use of quantile regression with continuous outcomes;
  • Become familiar with some of the background theory;
  • Become aware of some recent developments in quantile regression;
  • Be familiar with structure and content of SOEP survey data;
  • Apply the methods to real data;
  • Be trained in the use of statistical software for implementing the methods.

The course will use the STATA software. Prior familiarity with STATA would be an advantage although the course will cover some STATA basics. Familiarity with linear regression and basic distribution functions is assumed.

Requirement of students: intermediate statistical knowledge, basic Stata skills

Recommended literature and pre-readings:

  • Davino, C.; Furno, M. and Vistocco, D. (2014). Quantile Regression: Theory and Applications. John Wiley & Sons.
  • Haisken-DeNew, J.P. and Frick, J.R. (Eds.) (2005). DTC Desktop Companion to the German Socio-Economic Panel (SOEP). Version 8.0 – Dec 2005, Updated to Wave 21 (U). This documentation is intended to give novice users a “jump start” in understanding the SOEP, its structure, depth, and research potential.
  • Hao, L. and Naiman, D.Q. (2007). Quantile Regression. Quantitative Applications in the Social Sciences, No. 149.
  • Koenker R. (2005). Quantile Regression. Cambridge University Press.

You have to register for the 11th International Research Workshop to participate in this course.