Dr Yike Wang 32L.4.26, Prof Taisuke Otsu 32L.4.25, Prof Francisco Hidalgo 32L.4.20 and Prof Jorn Pischke 32L.2.16

This course is compulsory on the MRes/PhD in Finance. This course is available on the MRes/PhD in Economics, MRes/PhD in Economics and Management and MRes/PhD in Management (Marketing). This course is not available as an outside option.

Students should have completed an undergraduate level course in econometrics and statistical theory. Linear algebra and multivariate calculus will be used frequently.

This course consists of two parts, the core and topics parts. All students must follow the core part of the course for 13 weeks (MT Weeks 1-10 and LT Weeks 1-3), and then select between one of two streams for 7 weeks (LT Weeks 4-10).

The core part of the course begins with a review of linear regression analysis. It proceeds with discussions on linear instrumental variable (IV) regression, generalised method of moments (GMM), panel data analysis, nonparametric methods, and treatment effect analysis. Then it discusses estimation and inference on general nonlinear models including various limited dependent variable models. It also covers basics of time series analysis. Finally, additional topics such as bootstrap, quantile regression, and machine learning are also covered.

In the second part of the course, students select to be examined in one of two streams [though students may attend the lectures of both streams if they wish].

Stream 1 discusses various macroeconomic applications of econometrics methods covered in the core part, multivariate time series analysis including vector autoregression and impulse response analysis, Bayesian methods, and related computational methods. Then it discusses nonstationary time series, cointegration, inference with long memory data, nonlinear time series analysis including GARCH, stochastic volatility, and threshold models, and introduction to frequency domain analysis.

Stream 2 focuses on programme evaluation methods frequently used in applied microeconomics.  It discusses issues arising in regression control, instrumental variables, differences-in-differences and fixed effects methods, regression discontinuity designs, and statistical inference.  Throughout, the discussion are supported by many empirical applications.

30 hours of lectures and 15 hours of classes in the MT. 30 hours of lectures and 15 hours of classes in the LT.

30 hours of lectures and 15 hours of classes in the MT. 30 hours of lectures and 15 hours of classes in the LT. This course is delivered through a combination of classes and lectures totalling a minimum of 90 hours across Michaelmas Term and Lent Term. Attendance at lectures and classes is compulsory.

Compulsory exercises are set for each class.

Course material will be made available through the course Moodle page. Please note there is no set book for this course.

Recommended books are:

J. M. Wooldridge, Econometric Analysis of Cross Section and Panel Data, 2010

B. Hansen, Econometrics, 2022

J. Angrist and J. S. Pischke, Mostly Harmless Econometrics, 2009

F. Hayashi, Econometrics, 2000

T Amemiya, Advanced Econometrics, Harvard University Press, 1985;

P. J. Brockwell and R. A.Davis,  Time series: Theory and methods, 2006

W. A. Fuller, Introduction to Statistical Time Series, 1976.

Exam (100%, duration: 3 hours, reading time: 15 minutes) in the summer exam period.