Tuesday, November 29, 2005

MORE ILLUMINATING ANSWERS TO QUESTIONS FOR IMF ECONOMISTS


I write Nicoletta Batini, Senior Economist in the IMF Research Department and lead author of this chapter (*.pdf) in the latest World Economic Outlook:

Dear Nicoletta:

I have several questions with which I would be most grateful if you
could help me.

1. I want to ask if you could send me the data you used in the study?

2. The chapter is not quite clear on the countries used in the
non-inflation targeting control group. Aside from the 7 countries
identified in footnote 17, the chapter indicates you used the 22
remaining countries in the Morgan EMBI. By my count there should be 41
countries in the sample, is this right? It appears there are several
Morgan EMBIs, from which one are you srawing the sample?

3. Which period(s) are you looking at in the data? The text indicates
that the sample runs from 1985q1-2004q4, thought from the charts it
looks like
1990q1 - 2004q4. However observations are missing from the IFS data for
a number of sample countries in this time period for CPI, annual %
change in prices, and GDP.

4. Which measure of GDP did you use to test the effect of inflation
targeting on volatility of real output? Did you deflate by CPI or by the
GDP deflator? Similarly the quarterly GDP data are quite sparse in the IFS--especially for the seasonally adjusted data? If you used the un-SA data, might this be overstating the volatility of output (as well as inflation
volatility)?

Thank you in advance for your time and attention (and your important
work on inflation targeting). I look forward to hearing from you.

Sincerely,

Globalize This!


Nicoletta replies:

Dear GT,

Thanks for this email.

You find me in a very busy time so only a few quick answers.

We cannot release the data because of confidentiality.

Countries in the control group are the residual EMBI group, to date
(plus or minus the 7 tat are not EMBI included).

Charts sample and regression samples differ and don't need to agree...

Real GDP is GDP deflated.

Best

NB


Oh, well that clears it all up. No, wait a minute. That doesn't answer any of my questions. It's almost like she doesn't want me probing into her research.

0 Comments:

Post a Comment

<< Home