Spatial autocorrelation and extrapolation of purchasing power parities. Modelling and sensitivity analysis

Rambaldi, Alicia, Rao, D.S. Prasada and Ganegodage, K. Renuka (2009). Spatial autocorrelation and extrapolation of purchasing power parities. Modelling and sensitivity analysis. CEPA Working Paper Series WP01/2009, School of Economics, University of Queensland.


Author Rambaldi, Alicia
Rao, D.S. Prasada
Ganegodage, K. Renuka
Title Spatial autocorrelation and extrapolation of purchasing power parities. Modelling and sensitivity analysis
School, Department or Centre School of Economics
Institution University of Queensland
Series CEPA Working Paper Series
Report Number WP01/2009
Publication date 2009-03-18
Publisher The University of Queensland School of Economics
Start page 1
End page 31
Total pages 31
Language eng
Subject 140302 Econometric and Statistical Methods
140304 Panel Data Analysis
140202 Economic Development and Growth
C1
Abstract/Summary The paper examines the role and significance of modeling spatially correlated disturbances in the extrapolation of purchasing power parities (PPPs) within the general econometric framework developed by Rao et al for the purpose of constructing a consistent time-space panel of PPPs. Alternative measures of economic distance are considered using trade closeness as well as a constructed measure using a common factor approach which combines indicators of trade, cultural and geographical closeness. The measures are used to construct spatial weight matrices. A comparative analysis of the effect of alternative specifications of the spatialweight matrix on the PPP extrapolations is conducted with emphasis in the model's prediction ability. Specifically, the out-of-sample prediction of the PPPs for GDP recently released by the International Comparisons Program (World Bank) for the 2005 ICP Benchmark year are used to evaluate the alternative specifications. The results clearly indicate the need to model and use a spatially correlated error structure especially when the benchmark data are incomplete. The results are very similar when the spatial weight matrices are based on trade closeness or the more comprehensive economic distance measure.
Keyword Purchasing power parities
Spatial autocorrelation
Principal components
Economic distance
Kalman filter
Q-Index Status Provisional Code
Institutional Status UQ

 
Versions
Version Filter Type
Citation counts: Google Scholar Search Google Scholar
Created: Wed, 31 Mar 2010, 16:14:56 EST by Alys Hohnen on behalf of School of Economics