All Theses and Dissertations

Degree

Master of Science in Economics

Faculty / School

Faculty of Business Administration (FBA)

Department

Department of Economics

Date of Award

Spring 2017

Advisor

Ehsan U. Choudhri

Second Advisor

Dr. Adnan Haider

Committee Member 1

Ehsan U. Choudhri, Institute of Business Administration, Karachi

Committee Member 2

Dr. Adnan Haider, Institute of Business Administration, Karachi

Project Type

Thesis

Access Type

Restricted Access

Pages

xi, 95

Abstract

This study estimates a DSGE model and three versions of VAR models to analyze forecast performance of these models in context of Pakistan. Three versions of VAR models (VARX. BVARX and BVAR) and, a variant of Adolfson. Laséen. Lindé, and Villani (2007) DSGE model have been estimated using quarterly data (1980Q4-2016Q2). Expanding window recursive out-of-sample forecasts for GDP growth, call money rate, CPI inflation and change in exchange rate are generated and compared over the period 2009Q1-2016Q2. Forecasting performance is analyzed by the comparison of bias and root mean squared errors (RMSE). Comparison of forecasting performance shows that, VAR models provide better forecasts than estimated DSGE model. However, in case of GDP growth, interest rate and inflation, forecasting performance of estimated DSGE model appears to be quite close to VAR models. Forecasts from all models are positively correlated, detiorate in turbulant times and improve in relatively calm periods.

Download

The full text of this document is only accessible to authorized users.

Share

COinS