Matlab course for 1st year researchers (ECO-CO-MATLAB)

ECO-CO-MATLAB

Description

Instructor: Robert Schall, [email protected], Office: TBD

Prerequisites: Introductory classes in Mathematics (A. Villanacci) and Statistics (C. Lafuente)

Structure: The course consists of three parts. The first part covers the basics (objects, loops) required for more advances topics covered later in the course. It also addresses various concepts related to distributions, simulations, and random numbers. The second part introduces concepts specifically relevant for the computational exercises in Macroeconomics I and Macroeconomics II. In particular, this part is intended to teach Dynamic Programming using the Matlab. First, we will cover the cake eating problem and its extensions. Second, we will solve the Neoclassical Growth Model (Ramsey Problem).

Each session consists of two parts. During the first part (60-90 minutes), I will introduce the relevant concepts, Matlab commands, and coding examples as well as several smaller practical examples/exercises. During the second part of the session, we will work on a larger project that will allow us to apply the concepts learned.

I would suggest to do this last part of each session in an open-ended manner, such that everyone can do the project at their personal pace. Hence, each session will take up to 2:30h, including breaks.

Goal: By the end of the course, everyone should have a decent understanding of Matlab. In particular, you should (hopefully) be able to solve standard consumption-savings problems as encountered in Macroeconomics II (Cooper’s part) and solving the Neoclassical Growth Model using Value Function Iteration should be a familiar exercise to you. Jesus’ part of Macroeconomics I will take that as given.

PART I: Basics
Session 1: Introduction
Concepts: Starting Matlab, basic objects (Vectors, Matrices, Arrays,...), Operations, Logical statements,
Comparisons, Loops
Project: Simple SIR Model

Session 2: Distributions & Simulations
Concepts: Functions, (Plotting), Distributions & Random numbers, Maximum Likelihood Estimation (MLE)
Project: Estimating p using Monte Carlo Simulations

PART II: Dynamic Programming with Matlab
Session 3: Optimization & Dynamic Programming Intro
Concepts: Fixed point problems, Zero-finding, Minimization/Maximization (fminsearch etc.), Dynamic
Programming Introduction
Project: TBC: Two-period Optimization Problem (for instance, from Cooper’s Notes on Two Period
Optimization Problems or Villanacci script section 4.4) or Cake Eating Problem

Session 4: Dynamic Programming I
Concepts: Dynamic Programming, Uncertainty, Markov Chains
Project: Neoclassical Growth Model