GDP as the main indicator of the country's economic development is used by structures on the macro level for making managerial decisions. However, the official value of the Ukrainian GDP is published by the State Statistics Service with the quarterly periodic, so for decision-making between the publishing periods the forecasts are used. Such sort term predictions for the near feature or the resent past in economics are called nowcastings. The article implements the short-term forecast model of Ukraine's GDP based on ten indicators of socioeconomic condition of Ukraine, namely: real GDP, industrial output, capital investment, average wage, retail turnover, agricultural output, consumer price index and producer price index of industrial output, export and import of goods and services. The indicators are collected for the period from the first quarter of 2002 to the second quarter of 2019. Since the input statistics are collected with different frequency (GDP is a quarterly indicator, and, for example, consumer price index or average wage is disclosed monthly), the bridge model is applied, which combines variables of different frequency. In general, the main equation of the bridge model (bridge equation) can be represented as an autoregressive distributed lag model (ADL). Standard bridge models are in the form of simple linear equations, it makes it easier to understand the relationship between the included indicators and the forecast, but bridge models can include a small number of explanatory variables. By means of the procedure of automatic selection of the "PCGets" model based on the "from general to specific" modeling approach, the number of indicators for the quarterly GDP forecast is reduced from ten to four, namely, capital investment, imports, real household income and agricultural output. The forecast of the Ukrainian GDP for the next three quarters is estimated with the bridge model. The quarterly forecast of Ukraine's GDP in pseudo real time for the 2nd quarter of 2019 received with the bridge model deviates by 0.04% from the empirical value of GDP, which indicates the high quality of the forecast.
