The relationship between monetary growth and inflation: an application of the infinite hidden Markov model to the United Kingdom

Yu, Zhongning

Abstract

The analysis is undertaken through exploration of a reduced form relationship. Two questions are central to the study. The first involves the role of structural breaks in the relationship between inflation and monetary growth. We explore this through an application of Qu (2008) SQ and DQ tests which analyse structural breaks in both the mean and quantiles of the conditional distribution. The second question involves the role of nonlinearity in the relationship between inflation and monetary growth.

The results from the SQ and DQ tests suggest the existence of multiple structural breaks in the linear relationship between inflation and monetary growth. In the thesis, we propose a modification to the critical values underlying these tests to capture the effect of various sample sizes. The results of Monte Carlo experiments suggest that this modification improves the power of test when compared to the results given by Qu (2008).

From the estimation of Markov switch model and infinite Hidden Markov model, we find that the relationship between monetary growth and inflation exhibits a maximum of five regimes over the period 1966 to 2012. However, after introduction of inflation targeting in UK in 1992, the relationship between inflation and monetary growth stayed in one regime most of the time. The financial crisis of 2008 only changed the relationship between monetary growth and inflation for a short period before returning to the pre-crisis regime. The iHMM demonstrates a range of capabilities, notably the ability to detect structural change even at the end of the sample. This feature is desirable in monitoring potential structural breaks generally and given the importance of the specific relationship between money and inflation for practical policy purposes.

Citation

Yu, Z. The relationship between monetary growth and inflation: an application of the infinite hidden Markov model to the United Kingdom. (Thesis). Keele University. https://keele-repository.worktribe.com/output/410394

• Thesis Type: Thesis
• Publicly Available Date: May 26, 2023
• Keywords: nonlinearity, structural break, SQ test, DQ test, iHMM
• Public URL: https://keele-repository.worktribe.com/output/410394
• Award Date: 2018-03