This is a hands-on graduate-level course teaching key techniques to analyse and estimate macroeconomics models. It teaches the key building blocks of numerical analysis such as function approximation and numerical integration. The course shows how these techniques are used in perturbation and projection methods to accurately solve nonlinear dynamic stochastic models. Relevant theoretical aspects such as the Blanchard-Kahn conditions and the possibility of sunspots solutions are also covered. The course also teaches the tools to estimate such models (Kalman filter, Bayesian estimation, MCMC). Students are taught how to use Dynare, but also how to write Matlab programs to solve a variety of models with other techniques. In addition to teaching techniques, the course also focuses on practical problems that researchers run into when using these techniques.In the morning sessions, a lecture is given by one of the two instructors. In the afternoon sessions, students work in groups on computer assignments with the help of the instructors and teaching assistants.Monday - Solving and analysing your first DSGE model- State variables- Policy rules (i.e. the recursive solution to DSGE models)- Impulse response functions- Perturbation analysis- Certainty equivalence- Dynare- Using the homotophy idea to get good initial values for the steady state (often the hardest part of running Dynare)- Parameter values and properties of basic neoclassical model- Stylised factsTuesday - Key tools from the numerical approximation literature and projection methods- Numerical integration (Gaussian quadrature)- Function approximation (Splines & Polynomials)- Projection methods- Endogenous grid points- Fixed point iteration- Time iterationWednesday - Topics- Parameterized Expectations Algorithm- Value Function Iteration- Accuracy tests: Euler errors, Dynamic Euler equation test, DHM statistic- Occasionally binding constraints and penalty functions- Blanchard-Kahn conditions- Sunspots and self-fulfilling expectationsThursday - Kalman filter and full information methods- Kalman filter- State space form- Maximum Likelihood- Avoiding the singularity problemFriday - Bayesian estimation- Bayesian estimation- MCMC- Metropolis Hastings

Course leader
Professor Wouter den Haan (LSE)Dr Petr Sedlacek (Bonn)

Target group
The course is aimed at PhD students and academics.

Course aim
This course will aim to provide students with:- a chance to learn a solid set of different tools to analyse and estimate modern macroeconomic models;- a better understanding of the properties of modern macroeconomic models;- a better understanding of the importance of nonlinearities;- a better understanding of the limitations of popular techniques.

Fee info
GBP 725: Student rate - available to current university (including PhD) students. Academic staff and staff of UK charities are eligible for a reduced rate of 1,500. GBP 2300: Standard fee.

Scholarships
Current PhD students are also eligible for a 150 ESRC scholarship.

London School of Economics
Address: Houghton Street
Postal code: WC2A 2AE
City: London
Country: United Kingdom
Website: http://www.lse.ac.uk/study/summerSchools/Methods/home.aspx
E-mail: summer.methods@lse.ac.uk
Phone: +44 (0)20 7955 6422