Financial Econometrics

Syllabus

• Causal inference I: the potential outcomes model, its main elements and identifying assumptions; sufficient conditions for unbiased estimation of the average treatment effect.
• Causal inference II: classification of confounders, bad controls and related variables; examples, causal maps and omitted variable bias.
• Methods for addressing confounding: control variables, model specification and interpretation of causal effects.
• Experimental design: types of experiments, optimal treatment allocation, optimal stratification and the minimum sample size required to detect a given effect.
• Practical challenges in experiments: covariate balance, spillovers, missing data, Hawthorne effects, placebo effects and John Henry effects.
• Panel data methods: fixed-effect models, the Simpson paradox and clustered standard errors.
• Statistical illusions in empirical research: p-hacking, multiple hypothesis testing, family-wise error rate, Bonferroni correction and Holm's method.
• Regularization methods in econometrics: Lasso and adaptive Lasso.
• Granger causality and network approaches in time-series econometrics.
• Regularization for high-dimensional time-series models and econometric networks.
• Information criteria in high-dimensional time-series models.
• Bootstrap methods and their econometric interpretation.
• The Diebold-Yilmaz connectedness framework.
• The role of bootstrap methods in network-based econometric analysis.