Examples of Methods for Continuous Time Problems

General Lecture notes

• See Ben Moll's Heterogeneous Agent Models in Continuous Time project, and http://www.princeton.edu/~moll/notes.htm In particlar:
  • From Ben Moll: http://www.princeton.edu/~moll/Lecture1_Rochester.pdf and http://www.princeton.edu/~moll/Lecture2_Rochester.pdf
  • From Pontus Rendahl: http://www.princeton.edu/~moll/Pontus_Lecture1.pdf and http://www.princeton.edu/~moll/Pontus_Lecture2.pdf
  • Stopping Time Problem I: Exercising an Option::

Detailed Notes on Methods

• See operator_discretization_finite_differences.pdf for details on the discretization of the stochastic proceses (both stationary and time-varying), and the solution to simple value function and stationary distribution calculuations.
• See optimal_stopping.pdf for details on solving optimal stopping problem (given the discretization).
• See HJBE_discretization.pdf for details on solving the neoclassical growth model

Structure of folders

• The source for the documentation of all algorithms is in docs/.
• For the Matlab specific implementation of the algorithms
  • See library functions to implement the methods is in matlab/lib/
  • See examples which use the library functions in matlab/examples/. At this point, most of the code is in tests rather than examples.
  • All unit and regression tests of the matlab library are in matlab/tests/. You can use run_tests.m in the matlab folder to run all regression tests after making coding changes.