Discussion Paper

No. 2007-29 | July 10, 2007
Uncover Latent PPP by Dynamic Factor Error Correction Model (DF-ECM) Approach: Evidence from five OECD Countries


This study explores a new modelling approach to bridge the gap between the bilateral setting of one ‘domestic’ economy facing one ‘foreign’ entity in theory and multilateral country data in reality. Under the approach, purchasing power parity (PPP) is embedded in latent disequilibrium factors, being extracted from a large set of bilateral price disparities; the factors are then used as error-correction leading indicators to explain exchange rate and inflation. Modelling experiments on five OECD countries using monthly data show promising results, which reverse the common belief that PPP is at best a very long-run relationship at the macro level.


Paper submitted to the special issue "Recent Developments in International Money and Finance" edited by Ronald MacDonald

JEL Classification

C22 C33 F31

Cite As

Duo Qin (2007). Uncover Latent PPP by Dynamic Factor Error Correction Model (DF-ECM) Approach: Evidence from five OECD Countries. Economics Discussion Papers, No 2007-29, Kiel Institute for the World Economy. http://www.economics-ejournal.org/economics/discussionpapers/2007-29


Comments and Questions

Anonymous - Referee Report
February 11, 2008 - 10:24

see attached file

Anonymous - Reply to comments
February 12, 2008 - 16:20

I find the anonymous referrer's report extremely helpful. I've already become aware of a few problems that he/she points out, because I've carried out some further research and found ways of dealing with them. The other comments are very constructive and make me think more.
I'm definitely going to ...[more]

... revise the paper, hopefully in the near future.

Anonymous - Decision of the Associate Editor
March 06, 2008 - 10:16

Seen on the whole I would like to rate the paper as publishable in the if it is thoroughly revised according the lines sketched by Referee 1. Especially, the caveats raised under the heading "quality of analysis" have to be tackled. - I would like to see some reaction to ...[more]

... the fact that the authors remain completely silent on the implications of generated regressors in their multi-stage setting. Their approach with generated regressors ought to affect standard errors and test statistics at subsequent stages which definitely needs further methodological discussion. Moreover, the authors should carefully check whether the specification of the ECMs in the paper beginning in table 6 is of somewhat “hybrid” form. In a standard ECM conventionally testing for bilateral PPP, the change in a bilateral exchange rate would be regressed on an error correction term based on the bilateral PPP relationship. An analogous investigation should be performed on the basis of the effective exchange rate. However, in the ECM in the paper such as in table 6ff, the bilateral exchange rate with respect to the dollar is regressed on a multilateral (effective) exchange rate. Is the referee right in stating that the assessment of the power of the current methodology needs a conventionally specified ECM and not a “hybrid” one as a standard of reference? If yes, the authors should correct for it in the final version of the paper"

Duo Qin - Response List to the Reviewers Suggestions
March 06, 2008 - 10:21

see attached file

Duo Qin - Revised Version
March 06, 2008 - 10:24

sse attached file