Estimation of Uncertainty: Revisions of Russian GDP on History

Artur R. Sharafutdinov – Graduate Student of the Russian Presidential Academy of National Economy and Public Administration; Economist of the Monetary Policy Department, Bank of Russia (Moscow, Russia). Е-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.

The article is devoted to estimating the uncertainty parameters of Russian GDP on history, that arises as a result of revisions and refinements of data over time. A brief review of the reasons for the revisions allows us to form an understanding of their necessity and importance. For analysis, it is assumed that uncertainty decreases over time and occurs systematically, which is reflected in heteroscedasticity in the form of a decrease in the standard deviation of revisions over time and in autocorrelation in the form of a dependence of revisions within a single publication. Estimation is performed using a parametrized covariance matrix using the Cunningham method.

As an example of visualizing uncertainty in data, fan charts are presented in the work. One of the applications of historical uncertainty estimation is the ability to calibrate parameters to filter true GDP values, as well as the ability to refine forecast uncertainty.

Key words: Russian GDP revisions, GDP uncertainty parameters, vintage data analysis, fan chart plotting.

JEL-codes: C13, C30, C65.