Truths and myths about RMB misalignment : A meta-analysis
Cheung, Yin-Wong; He, Shi (22.02.2019)
Numero
3/2019Julkaisija
Bank of Finland
2019
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:bof-201902281072Tiivistelmä
We conduct a meta-regression analysis of 69 studies that generated 937 renminbi (RMB) misalignment estimates. The Bayesian Model Averaging (BMA) approach is adopted to allow for model selection and sampling uncertainties in assessing effects of study characteristics on these RMB misalignment estimates. Misalignment estimates are found to be influenced by the eight selected study characteristic types in our median probability model. The RMB misalignment estimate from models with various hypothetical combinations of study characteristics, however, is mostly insignificantly different from zero. It is also shown that the set of significant study characteristics is sensitive to the use of the least squares estimation method and the choice of benchmark study characteristics.
Sisällysluettelo
Abstract 4
1 Introduction 5
2 Preliminaries 7
2.1 Sample of studies 8
2.2 Study characteristics 9
2.2.1 Data characteristics 9
2.2.2 Theoretical and estimation specifications 10
2.2.3 Publication attributes 11
2.2.4 Subsample periods 12
3 Data analyses 12
3.1 Basic BMA results 14
3.2 Misalignment under hypothetical combinations of study characteristics 17
4 Additional analyses 19
4.1 Regression analysis 19
4.2 Alternative BMA results 20
5 Concluding remarks 23
References 24
Tables and figures 27
Appendices 39
A1 Sample of studies 39
A2 Percentage misalignment estimates: Descriptive statistics 41
A3 Study characteristic types 42
A4 Bayesian Model Averaging 44
1 Introduction 5
2 Preliminaries 7
2.1 Sample of studies 8
2.2 Study characteristics 9
2.2.1 Data characteristics 9
2.2.2 Theoretical and estimation specifications 10
2.2.3 Publication attributes 11
2.2.4 Subsample periods 12
3 Data analyses 12
3.1 Basic BMA results 14
3.2 Misalignment under hypothetical combinations of study characteristics 17
4 Additional analyses 19
4.1 Regression analysis 19
4.2 Alternative BMA results 20
5 Concluding remarks 23
References 24
Tables and figures 27
Appendices 39
A1 Sample of studies 39
A2 Percentage misalignment estimates: Descriptive statistics 41
A3 Study characteristic types 42
A4 Bayesian Model Averaging 44
Julkaisuhuomautus
Published in Comparative Economic Studies Volume 61, Issue 3, September 2019, Pages 464–492 https://doi.org/10.1057/s41294-019-00093-0