Nowcasting Ukraine's GDP using a factor-augmented VAR (FAVAR) model
Journal Title: Visnyk of the National Bank of Ukraine - Year 2017, Vol 0, Issue 242
Abstract
This article presents an approach for nowcasting the current value of Ukraine’s quarterly GDP. The approach uses leading indicators with a different disclosure frequency. We generalize data from a set of explanatory variables into several factors by using principal components analysis and estimate the factor-augmented VAR (FAVAR) model. Our system incorporates new data as they are published throughout a quarter to adjust GDP nowcasts. In addition, we research the influence of separate data releases on the accuracy of forecasts.
Authors and Affiliations
Anton Grui, Roman Lysenko
Keywords
Related Articles
- EP ID EP426662
- DOI 10.26531/vnbu2017.242.005
- Views 134
- Downloads 0
The NBU’s Balance Sheet: Before, During, and After the Crisis
Looking at the evolution of the central bank’s balance sheet gives us a unique window on the forces that have shaped our economy and central bank reaction functions. This paper considers the evolution of the NBU’s balanc...
What Drives the Difference between Online and Official Price Indexes?
This paper examines the associations between online price indexes and official statistics. First, we generate online CPI component sub-indexes, which are later aggregated to an Online Price CPI. This approach is applied...
A Neutral Real Interest Rate in the Case of a Small Open Economy: Application to Ukraine
This paper measures a neutral interest rate in Ukraine by means of applying a Kalman filter to a semistructural model with unobserved components. We rely on a medium-term concept of a neutral interest rate, where it is d...
Independence of Central Banks in Commodity Economies
This article introduces the hypothesis that resource-rich countries display a low degree of central bank independence (CBI). This hypothesis is proven based on multivariable regression, but the influence of resource fact...
Ukrainian Banks’ Business Models Clustering: Application of Kohonen Neural Networks
Abstract This paper clusters and identifies six distinct bank business models using Kohonen Self-Organising Maps. We show how these models transform over the crisis and conclude that some of them are more prone to defau...