Resource Center

Advanced Search
Technical Papers
Working Papers
Research Memoranda
GTAP-L Mailing List
CGE Books/Articles
Important References
Submit New Resource

GTAP Resources: Resource Display

GTAP Resource #4132

"Bayesian Estimation of Input-Output Tables for Russia"
by Potashnikov, Vladimir, Oleg Lugovoy and Andrey Polbin

Russian statistical system is under transition for almost two decades from Soviet type Material Product System to SNA. The main transitional break in methodology took place in 2004-2005 when Russian statistical agency “Rosstat” started reporting based on the new definition of economic sectors consistent with NACE§, and stopped reporting using definition of activities inherited from the Soviet statistical system. This methodological break splits all industry level statistics into two periods with little consistency between each other. As a result, Rosstat stopped updating IOT in 2003, based on the only benchmark survey conducted in 1995. The next survey is scheduled for 2012 with expected publication of results in 2015 or later. Official backward estimation is not expected. Therefore Russian statistics will miss IOT at least from 2004 to 2010. Also quality of officially updated IOT from 1996 to 2003 based on 1995 benchmark is questionable.
The paper pursues several goals. First, to apply Bayesian statistics for updating IO tables for 1996-2004 period, i.e. within “old” definition of industries. Second, we apply Monte Carlo Markov Chains (MCMC) methods to disaggregate available in NACE classification SUTs (2004 to 2006, 15 products by 15 activities) into larger 69 by 69 format. Third, since the 15x15 SUTs are published by Rosstat as preliminary estimates, they are not fully consistent with other available national accounts data, such as output and value added by industries. To take into account the data uncertainty, we introduce a measurement error for the aggregated io-coefficients. As result, we estimate posterior distribution of input-output coefficients for aggregated and disaggregated matrices, which are consistent with yearly national accounts information. Forth, we update the estimated 15x15 and 69x69 matrices for 2007-2011 period using proposed sampling methods. Also we discuss potential ways to improve the estimates imposing additional constrains and li...

Resource Details (Export Citation) GTAP Keywords
Category: 2013 Conference Paper
Status: Published
By/In: Presented at the 16th Annual Conference on Global Economic Analysis, Shanghai, China
Date: 2013
Created: Potashnikov, V. (4/14/2013)
Updated: Potashnikov, V. (4/14/2013)
Visits: 1,924
- Calibration and parameter estimation
- Other data bases and data issues

If you have trouble accessing any of the attachments below due to disability, please contact the authors listed above.

Public Access
  File format GTAP Resource 4132  (616.0 KB)   Replicated: 0 time(s)
  File format GTAP Resource 4132  (616.0 KB)   Replicated: 0 time(s)

Restricted Access
No documents have been attached.

Special Instructions
No instructions have been specified.

Comments (0 posted)
You must log in before entering comments.

No comments have been posted.