Authors: Roman Lysenko, Natalia Kolesnichenko.
Abstract: This study aims to investigate to which extent the survey of Ukrainian firms’ business expectations can be used to forecast changes in real GDP and two of its components – consumption and investment. Enterprise survey results are aggregated using the principal component method. An alternative indicator is the Business Expectations Index, which is calculated by the National Bank of Ukraine. The GDP and its components are forecasted using ARDL and VAR models, which include calculated main components, the index of business expectations, and answers of firms regarding investment into construction in the next 12 months. We use pseudo-out-of-sample prediction to estimate the model's predictive power. Comparison of projected results shows that GDP growth and consumption in annual terms are best estimated using the results of enterprise surveys aggregated by the PCA and the first difference of the business expectations index.
Cite as: Lysenko, R., Kolesnichenko, N. (2018). Use of business expectations for short-term forecasting of economic activity in Ukraine. NBU Working Papers, 2/2018. Kyiv: National Bank of Ukraine. Retrieved from https://bank.gov.ua/admin_uploads/article/wp_nbu_2018-2_Lysenko.pdf.