| Kredit: | 2 |
|---|---|
| Role předmětu: | Anglicky CSF - elective Doktorský EEI a HP - povinně volitelný ET - povinně volitelný F,FT a B - povinně volitelný Magisterský - vše MEF - elective NEVYUČUJE SE Semestr - letní |
| Garanti: | doc. PhDr. Jozef Baruník Ph.D. Prof. Bryan S.Graham |
| Stránky kurzu: | JEM210 |
| Literatura: | |
| Popis: | This course will introduce students to some recent methods for analyzing network data. Methods of storing, displaying and summarizing network dataset will be reviewed. After this warm-up, methods of parametric and non-parametric dyadic regression will be comprehensively introduced. Dyadic data arises frequently in economics, canonical examples are trade and migration flows between nations. We will learn how to fit, and conduct inference on the parameters of gravity models of trade (for example) in a way which fully accounts for dyadic dependence. Some other methods will be presented in brief. Empirical examples and computational illustrations (using Python 3) will be featured throughout the course. Recommendation: A good masters or Ph.D. level year long introductory econometrics sequence; familiarity with probability and inference at the level of, for example, Casella and Berger (2001, Statistical Inference); a good understanding of linear regression and maximum likelihood estimation (especially as applied to binary choice models) will be assumed. Some exposure to non-linear panel data and U-Statistics is useful, but neither expected nor required. Efforts will be made to accommodate students with varying backgrounds by incorporating a good mix of elementary as well as more advanced material |
