Overview

The lectures provide a self contained introduction to the building, simulation and estimation macroeconometric models that constitute the main workhouse of today’s macroeconomics. These models are now commonly used in central banks for policy projections and business cycle analysis. Their interest is to mix micro foundations, dynamic relations and rational expectations. Furthermore, they can be estimated using recent developments in Bayesian econometrics. The objective of this course is to provide the both basic and advanced tools to allow participants to get some intuitions on these models and to be able to read and understand policy and research papers using this approach. Lectures are followed by practical sessions with hands-on computational exercises.

Handouts list

Note that this list is preliminary. Still not yet finished to upload all the content of the lecture.

Handout 1: Modern Business Cycle Theory
Objectives of the Lecture:
  • Understanding the concept of a business cycle;
  • Extracting the business cycle component from time series;
  • Solving a simple rational expectation model;
  • Linearizing a system of equations;

Materials:

chapter1.pdf Handout
log vs. level GDP vs log GDP (MATLAB file)
bpass.m Bandpass filter (MATLAB function)
BSmeasures.m Extract the business cycle component (MATLAB file)
Handout 2: A Production Economy
Objectives of the Lecture:
  • Solving a rational expectation model;
  • Linearizing a system of equations;
  • Extend a model.

Materials:

chapter2.pdf Handout
basicRBC.mod 3 equation RBC (Dynare file)
compareRBC.mod Solutions comparison (Dynare file)
Handout 3: Maximum Likelihood Estimation
Objectives of the Lecture:
  • Understanding how to use filters to obtain the sequence of errors;
  • Constructing the likelihood function based on the filter’s output;
  • Optimizing a likelihood function;
  • Applying these concepts via Dynare.

Materials:

chapter3.pdf Handout
LK.m Compare likelihood functions (MATLAB file)
MLE.m MLE estimation on gaussian variable (MATLAB file)
invertRBC.m DSGE-MLE estimation with inversion filter (MATLAB file)
kalmanRBC.m DSGE-MLE estimation with Kalman filter (MATLAB file)
KF.m Kalman filter function (MATLAB file)
KalmanLLK.m Likelihood function with Kalman filter (MATLAB file)
mydb.m Load macro-data (MATLAB file)
RBC_estim.mod Estimation of the RBC model (Dynare file)
Handout 4: Bayesian Estimation of DSGE Models
Objectives of the Lecture:
  • Estimating a DSGE model using Bayesian techniques;
  • Decomposing business cycles into multitple sources of exogenous disturbances;
  • Provide some forecasting analysis.

Materials:

chapter4.pdf Handout
bayesianRBC.m Estimate the error variance of a RBC Model w/o Dynare (MATLAB file)
RBC_estim.mod Estimate with Dynare the simple RBC model (Dynare file)
mydb.m Download observable variables of RBC model (MATLAB file)
fullRBC.mod RBC Model (Dynare file)

Get MATLAB on your computer


Two possibibilities
Please use your own laptop if you can with MATLAB installed prior to the first lecture.
1/ Installing: Dauphine University provides a free licence for students. Please have Datafeed, Optimization and Econometrics toolboxes installed. This is the safest solution
2/ Citrix: Go to Citrix to connect to a remote server hosted in Dauphine. Note that your internet connection must be stable, which may not always be the case if many users are simultaneously connected to the wifi.

Real time macro-data on MATLAB


This is a must-read as all of the lectures will rely on DBnomics. The latter is a database aggregator which can be queried directly from MATLAB. Please read carefully this following note on DBnomics and download as well the MATLAB function which allows to get real time macro data:

DBnomics.pdf
Quick Guide.     call_dbnomics.m
Query DBnomics w/ MATLAB.

The Dynare package


How to set up Dynare
1/ Download link. (For those of you using citrix, I suggest you to use an older version of dynare because of hard drive limitation, install on H:/ drive.)
2/ Load automatically Dynare.
3/ Call dynare by typing in the command window of MATLAB “dynare NAME_OF_YOUR_DYNARE_FILE”.

Toy models list

For your assessment, you must pick one of these toy models and estimate it with Bayesian techniques and dbnomics.

Involuntary unemployment

Materials:

unemployment.pdf Unemployment toy model
unemployment_RBC.mod unemployment toy model (Dynare file)
Open economy

Materials:

two_countries.pdf 2 country toy model
soe_RBC.mod 2 country toy model (Dynare file)
Mortgage market

Materials:

mortgage_market_model.pdf 2 country toy model
mortgage_RBC.mod Mortgage toy model (Dynare file)
Corporate loans market

Materials:

credit_market_model.pdf Firms credit frictions toy model
credit_RBC.mod Corporate loans toy model (Dynare file)