Course
Multivariate analysis: applied factor and regression analysis (PHD104)
The aim of this course is to provide PhD researchers with the necessary skills to produce publishable quantitative analyses. The course covers basic programming in R, as well as key topics in statistical modeling and inference. The course emphasizes careful identification of causal effects and proper modeling of different data generating processes, although the main focus will be on basic linear and generalized linear models. Rather than covering all techniques that could be relevant to a PhD project, the goal is to provide a set of skills that enable researchers to identify and apply more specific methods on their own.
Dette er emnebeskrivelsen for studieåret 2019-2020. Merk at det kan komme endringer.
Semesters
Fakta
Emnekode
PHD104
Vekting (stp)
5
Semester undervisningsstart
Spring
Undervisningsspråk
English
Antall semestre
1
Vurderingssemester
Spring
Content
Learning outcome
Forkunnskapskrav
Anbefalte forkunnskaper
Exam
| Form of assessment | Weight | Duration | Marks | Aid | Exam system | Withdrawal deadline | Exam date |
|---|---|---|---|---|---|---|---|
| Paper (6-10 pages) | 1/1 | Passed / Not Passed | — | — | — |
Each participant will write a short paper on a self-chosen topic. This task includes picking an answerable question, finding relevant data, and conducting an appropriate quantitative analysis. Complete R codes leading from original data to final results should be included as an appendix. In so far as possible, participants are asked to pick a topic of relevance to their ongoing research. The papers should be submitted within six weeks after the course has finished.
Coursework requirements
- Presence and active participation throughout the whole week.
- A one-page note on a tentative topic for the course paper, outlining potential data sources and methods. This note must be submitted one week in advance of the course.
- An individual presentation at the end of the course week, elaborating on the previously submitted note and outlining the course paper.
Fagperson(er)
Course coordinator:
Espen OlsenCourse teacher:
Espen OlsenMethod of work
The course will consist of one intensive week of lectures and applied seminars, Monday through Friday, from 10.15 to 15.00. Participants are expected to review lecture materials on their own.
The relevant software (R and Rstudio) is free and open source. Participants will need to bring well-functioning laptops on which the software can be installed. Participants who use laptops supplied by their university should install the software in advance (and contact IT for administrator rights if necessary).
Åpent for
PhD candidates enrolled in PhD programmes at the University of Stavanger or accredited universities/university colleges in Norway or abroad.
Single Course Admission to PhD-Courses.
Emneevaluering
Litteratur
Arnold, J. B. (2019). R for data science: Exercise solutions. [Available for free at: https://jrnold.github.io/r4ds-exercise-solutions/https://jrnold.github.io/r4ds-exercise-solutions/ ]
Gelman, A., & Hill, J. (2007). Data analysis using regression and multilevel/hierarchical models. Cambridge university press. [Key chapters: 3, 5, 6, 8, 9, 11, 12, 25. This book is available electronically. ]
Kim, J. O., & Mueller, C. W. (1978). Introduction to factor analysis: What it is and how to do it. Sage University Paper Series on Quantitative Applications in the Social Sciences (No. 13). Sage.
Wickham, H., & Grolemund, G. (2016). R for data science: import, tidy, transform, visualize, and model data. O'Reilly Media, Inc. [Updated version available for free at: https://r4ds.had.co.nz/https://r4ds.had.co.nz/ ]
Wooldridge. J. M. (2013). Introductory econometrics: A modern approach. 5th ed. South-Western, Cengage Learning. [Only Appendix C1-C3. This material is available electronically.]
Note: R for Data Science (Wickham et al. 2016) is available for free online (see the link), but can optionally be bought in print. This spring (2020), Gelman, Hill, and Vehtari are releasing the first volume of a major update to Gelman and Hill (2007), called "Regression and Other Stories". This volume covers the more basic parts of the old book, and it is better structured. It is also more up to date, especially when it comes to Bayesian estimation. Those who are interested in buying the course literature are advised to buy the new book rather than Gelman and Hill (2007). The latter includes a few more advanced topics, but it is available electronically and can be used to supplement the more recent volume where necessary. Minor electronic sources may still be added to the reading list.