Rational expectations versus adaptive learning in a DSGE model

Crawford School of Public Policy | Centre for Applied Macroeconomic Analysis

Event details

Seminar

Date & time

Tuesday 16 December 2014
12.00pm–1.00pm

Venue

Seminar Room 2, Crawford School of Public Policy, #132 Lennox Crossing, ANU

Speaker

Angelia Grant, PhD student, Centre for Applied Macroeconomic Analysis, Crawford School, ANU.

Contacts

Rossana Bastos
6125 8108

In this seminar, Angelia will provide an overview of her recent paper ‘Rational Expectations versus Adaptive Learning in a DSGE Model’. This paper replaces the assumption of rational expectations with adaptive learning in a widely cited DSGE model and compares the implied expectations generated by the different models with actual and survey data. It also assesses the overall relative fit of the models with the deviance information criterion, which is not commonly used to compare DSGE models. It is found that the assumption of rational expectations performs well. It provides an overall model fit that is comparable to the adaptive learning models, and is comparable in fitting implied expectations to actual consumption and inflation data and expectations from the Survey of Professional Forecasters. The adaptive learning models provide a slightly better fit to actual and survey investment data.

Angelia is a third year PhD student in Economics at CAMA. She was awarded a Sir Roland Wilson Foundation Scholarship to undertake research on business cycles and economic fluctuations, with a particular focus on comparing conclusions based on different economic models. The Sir Roland Wilson Foundation provides scholarships to public servants to undertake research that is of direct and enduring public policy relevance.

The CAMA Macroeconomics Brown Bag Seminars offer CAMA speakers, in particular PhD students, an opportunity to present their work in progress in front of their peers, and reputable visitors to showcase their work.

Updated:  18 July 2024/Responsible Officer:  Crawford Engagement/Page Contact:  CAMA admin