Till startsida
To content Read more about how we use cookies on gu.se

Advanced Regression Techniques (7.5 HEC) - Spring 2019

This course is a quantitative course at postgraduate level and is given as an elective course for students in the Faculty of Social Sciences. The course consists of four sections that are examined separately. The teaching of the course uses lectures, seminars and several 'applied workshops' in which students will use and apply what has been discussed in the course of data analysis in STATA. R-syntaxes will also be available.

The first section of the course is Multiple Regression Analysis and Logistic Regression Analysis (2 credits) which includes a linear regression analysis (OLS regression) with several explanatory variables (independent variables) and the management of dichotomous dependent variables (e.g., vote for candidate ‘A’ or not, a country-year with war or not, etc.). Specification of models, regression diagnostics, interaction effects and assumptions are all taken up in the readings and lectures.

The second subsection, Multilevel Analysis (2 credits), focus on when the data is on several levels (e.g. municipal, entity, individual). Specification of the basic models, estimation of parameters, model comparisons and assumptions are discussed here.

The third subsection, Panel Data Analysis (2 credits), focus on various techniques for analysis of data observed over time (e.g. studying changes and dynamic processes within units of analysis). Specification of models, correction of error, causality and assumptions are discussed here.

Finally, the fourth subsection, Causality and Instrumentation (1,5 credits), focus on causal analysis and statistical inference in terms of measurement errors, model specification and omitted variable bias. The section covers the use and implementation of instrumental variables and two-stage least squares estimation.

Generally, this is a course at intermediate level, which means that a pre-requisite is that you possess a basic knowledge of statistics and quantitative methods corresponding to the type of teaching that is usually taught in the undergraduate and graduate levels.

The examinations are conducted by four individual take-home exams.

The course starts on Monday, 15th January (week 3) and runs at half speed (daytime) until Wednesday 29th March. The course is taught by teachers from different institutions including political science, psychology and economics. All lectures will take place between the hours: 13:15 to 16:00 in room D207 on Sprängkullsgatan 23.


Date, Content and Atlas

week 3-4:
Introduction to the course & introduction to STATA.
Multiple Regression Analysis
Atlas: 15, 22, 24 & 26 Jan

week 5-6:
Logistic Regression
Atlas: 2, 5, 7 Feb *take-home exam for moment 1 is from 12-14 Feb

week 8-9:
Multilevel Analysis
Atlas: 19, 21 & 26 Feb *take-home exam for moment 2 is from 28 Feb-2 Mar.

Week 10-11:
Panel Data Analysis
Atlas: 5, 7 & 9 March *take-home exam for moment 3 is from 12-14 March

Week 12-13:
Causality and Instrumentation
Atlas: 19, 21 & 26 March *take-home exam for moment 4 is from 28-30 March

Teachers in the course sections:

1. Nicholas Charron, Associate Professor, Department of Political Science: nicholas.charron@pol.gu.se - Multiple and Logistic Regression.
2. Valgeir Thorvaldsson, Associate Professor, Department of Psychology: valgeir.thorvaldsson@psy.gu.se - Multilevel Analysis.
3. Stefan Dahlberg*, Associate Professor, Department of Political Science: stefan.dahlberg@pol.gu.se - Panel Data Analysis (*responsible for the course).
4. Alpaslan Akay, Associate Professor, Department of Economics: alpaslan.akay@gu.se - Causality and Instrumentation


As soon as possible, but by Dec 13 at the latest, please contact: lena.caspers@gu.se.
Please include your name, ’personnummer’, and to which institute you are affiliate as a gradutate student at GU. Please include your preferred language of instruction (English or Swedish)




Page Manager: Karl-Fredrik Ahlmark|Last update: 8/29/2018

The University of Gothenburg uses cookies to provide you with the best possible user experience. By continuing on this website, you approve of our use of cookies.  What are cookies?