Konstrukce a využití zrychleného odhadu čtvrtletního vývoje hrubé přidané hodnoty České republiky

Estimates of gross domestic product (GDP) and gross value added (GVA) belong among the most important characteristics of the development of the economic situation of each developed country. They are published with the task of meeting users’ demand for the fastest information on the current developments in the economy. There is a great deal of attention laying on the first published quarterly estimate of GDP, so-called flash estimate, as it provides the first signal within the system of national accounts with a major influence on the decision-making by the central authorities. Czech Statistical Office publishes flash estimate approximately 45 days after the end of the reference quarter. Based on its recommendations, Eurostat requires the EU countries to publish their flash estimates within 45 to 60 days after the end of the reference quarter. As a result of growing demand for the most up-to-date information on economic developments, it also focuses on the estimates available 30 days after the end of the reference quarter. The aim of this thesis is to assess the existing possibilities of constructing earlier estimate, comparing individual approaches and selecting the most suitable model for the construction of rapid estimate of GVA’s development as an earlier alternative to the published flash estimate. The theoretical part provides an overview of domestic and international experience with the construction of flash estimates and methods used in this thesis. Furthermore, the data set as the basis for the practical part is presented. Practical part is devoted to the construction of the rapid estimate using quarterly and monthly indicators related to the industries significantly contributing to the GDP formation. The benefit of the thesis is represented by the introduction of a comprehensive overview of the methods commonly used in practice to estimation of the quarterly GVA development and their application to the official data for the Czech Republic. This connection makes possible to compile earlier estimate of GVA growth available up to three weeks earlier than the flash estimate. Such an earlier estimate (referred to as rapid estimate) reaches quite satisfactory results within the test period, although the verification process and subsequent ex post analysis have revealed partial shortcomings. Their future resolving can contribute to the improvement of this earlier estimate. Special attention should be paid to the manufacturing industry where the implemented methods do not achieve optimal results.