Business cycles, interest rates and market volatility : estimation and forecasting using DSGE macroeconomic models under partial information

Perendija, Djordje V. (2018) Business cycles, interest rates and market volatility : estimation and forecasting using DSGE macroeconomic models under partial information. Doctoral thesis, London Metropolitan University.


Even long before the recent financial and economic crisis of 2007/2008 economists were more than aware of the insufficiencies and a lack of realism in macroeconomic modelling and model calibration methods, including those with DSGE methods and models, and spelled the need for further enhancements. The issues this research started addressing even before the 2008 crisis imposed demand for improvements, was use of single, fully informed rational agents in those modes. Consequently, the first part of this research project was aiming to improve the DSGE econometric methods by introducing novel solution for DSGE models with imperfect, partial information about the current values of deep variables and shocks, and apply this solution to imperfectly informed multiple agents with their different, inner-rationality models. Along these lines, this research also shows that DSGE models can be extended and suited to both, fitting and estimation of long-term yield curve, and to estimating with rich data sets by extending further its inner-mechanism.

In the aftermath of the 2008 crises, which struck at the beginning of this research project, and the subsequent, extensive criticism of DSGE models, this research analyses the alternative causes of the crisis. It then focuses on identifying its possible causes, such as yet unknown debt accelerator mechanism and the related, probable model miss-specifications, rational inattention, and as well, a role of institutional policies in both the development of the crisis and its resolution.

And finally, in a response to many of the critiques of the, usually monetary policy oriented DSGE models, this research project provides another set of novel extensions to such models, aiming to bring more of Keynesian characteristics suited to a more active, endogenous fiscal policy deemed needed in the aftermath of the crisis. This project, henceforth, extends the NK-Neo-Classical synthesis monetary DSGE models with a novel, endogenous, counter-cyclical fiscal policy rule driven by news and unemployment changes. It then also shows overall benefits of the resulting, mutually active, monetary-fiscal policy for both capital utilisation and overall economic stability.

PerendijaGeorge(Djordje)_Thesis.pdf - Published Version

Download (2MB) | Preview


Downloads per month over past year

Downloads each year

View Item View Item