Nowcasting Aggregate Financial Indicators of Russian Enterprises
Abstract
The research is devoted to constructing models that are able to nowcast (assess the current state) aggregate financial indicators of Russian organizations, the official statistics on which are becoming available with a delay. Sixteen target variables, which include profits of enterprises, current and non-current assets, revenue, expenses of various categories, indicators of total and overdue debt, are considered across various sectors: agriculture, mining, manufacturing, construction, trade and the whole economy. Quickly published economic and financial indicators of the Russian economy, including ones that take into account industry specifics, are used as predictors. Attention is focused on constructing interval (probabilistic) nowcasts that reflect a more complete picture of variables’ dynamics and support risk assessment. Despite the abundance of domestic studies on the nowcasting of macroeconomic indicators, including the ones involving probabilistic forecasting methods, the problem of prompt assessment of aggregated financial indicators has not been previously addressed. Linear regression, quantile regression, and quantile regression neural network are used as the forecasting tools. Predictors are selected using linear and quantile LASSO-regressions. The models are compared with first-order autoregression and dynamic factor model. The optimal model is selected for each target variable after testing on historical data. According to the results obtained, different tools should be used to produce nowcasts of selected indicators. Nevertheless, neural network, which has the advantage of modeling complex nonlinear dependencies, turns out to be the best approach to interval and point nowcasting for the largest number of targets. The constructed models can be used by public authorities to obtain prompt data on financial indicators and develop timely measures to support Russian enterprises, as well as in the framework of tasks such as assessing the debt burden and financial stability of business, planning budget revenues and tax policy.
Downloads
References
Adrian T., Boyarchenko N., Giannone D. (2019) Vulnerable Growth. The American Economic Review, 109, 4, pp. 1263–89. DOI: https://doi.org/10.1257/aer.20161923
Cannon A. (2011) Quantile Regression Neural Networks: Implementation in R and Application to Precipitation Downscaling. Computers & Geosciences, 37, 9, pp. 1277–1284. DOI: https://doi.org/10.1016/j.cageo.2010.07.005
Carriero A., Clark T.E., Marcellino M. (2020) Nowcasting Tail Risks to Economic Activity with Many Indicators. Working Paper no 20–13R2. Federal Reserve Bank of Cleveland. DOI: https://doi.org/10.26509/frbc-wp-202013
Carriero A., Clark T.E., Marcellino M. (2022) Specification Choices in Quantile Regression for Empirical Macroeconomics. Working Paper no 22–25. Federal Reserve Bank of Cleveland. DOI: https://doi.org/10.26509/frbc-wp-202225
Fokin N.D. (2023) Nowcasting and Forecasting Key Russian Macroeconomic Variables With the MFBVAR Model. Economic Policy, 18, 3, pp. 110–135. (In Russ.) DOI: https://doi.org/10.18288/1994-5124-2023-3-110-135
Gareev M.Y., Polbin A.V. (2022) Nowcasting Russia’s key Macroeconomic Variables Using Machine Learning. Voprosy Ekonomiki, 8, pp. 133–157. (In Russ.) DOI: https://doi.org/10.32609/0042-8736-2022-8-133-157
He Y., Li H. (2018) Probability Density Forecasting of Wind Power Using Quantile Regression Neural Network and Kernel Density Estimation. Energy Conversion and Management, 164, 15, pp. 374–384. DOI: https://doi.org/10.1016/j.enconman.2018.03.010
He Y., Qin Y., Wang S., Wang X., Wang C. (2019) Electricity Consumption Probability Density Forecasting Method Based on LASSO-Quantile Regression Neural Network. Applied Energy, 233–234, pp. 565–575. DOI: https://doi.org/10.1016/j.apenergy.2018.10.061
He Y., Xu Q., Wan J., Yang S. (2016) Short-Term Power Load Probability Density Forecasting Based on Quantile Regression Neural Network and Triangle Kernel Function. Energy, 114, 1, pp. 498–512. DOI: https://doi.org/10.1016/j.energy.2016.08.023
Kazakova M.V., Fokin N.D. (2024) Testing Forecasting Properties of Different Approaches to Interval Forecasting (Using the Example of Inflation in Russia). Voprosy Statistiki, 31, 5, pp. 23–40. (In Russ.) DOI: https://doi.org/10.34023/2313-6383-2024-31-5-23-40
Koenker R. (2005) Quantile Regression. Cambridge University Press. DOI: https://doi.org/10.1017/CBO9780511754098
Kryzhanovskiy O., Mogilat A., Shuvalova Zh., Gvozdev D. (2025) Using LSTM Neural Networks for Nowcasting and Forecasting GVA of Industrial Sectors. Russian Journal of Money and Finance, 84, 1, pp. 93–104.
Lyakhnova M., Kolenko Y. (2024) Nowcasting the Output Gap in Russia Using Enterprise Monitoring Data. Russian Journal of Money and Finance, 83, 2, pp. 26–53.
Maiorova K., Fokin N. (2021) Nowcasting Growth Rates of Russia’s Export and Import by Commodity Groups. Russian Journal of Money and Finance, 80, 3, pp. 34–48. DOI: https://doi.org/10.31477/rjmf.202103.34
Makeeva N., Stankevich I. (2022) Nowcasting of the Components of Russian GDP. HSE Economic Journal, 26, 4, pp. 598–622. (In Russ.) DOI: https://doi.org/10.17323/1813-8691-2022-26-4-598-622
Makeeva N.M., Stankevich I.P., Lyubaykin N.S. (2024) Nowcasting the Russian Economy Macroeconomic Indicators under Uncertainty: Does Taking into Account the News Sentiment Help. Voprosy Ekonomiki, 3, pp. 120–142. (In Russ.) DOI: https://doi.org/10.32609/0042-8736-2024-3-120-142
Makeeva N. (2025) The Impact of the Official Statistics Revision on the Accuracy of the Russian Macroeconomic Indicators Nowcasting Models. Applied Econometrics, 79, pp. 27–49. (In Russ.)
Mitchell J., Poon A., Mazzi G.L. (2022) Nowcasting Euro Area GDP Growth Using Bayesian Quantile Regression. Essays in Honor of M. Hashem Pesaran: Prediction and Macro Modeling (eds. A. Chudik et al.). Emerald Group Publishing Limited. Ch. 3. DOI: https://doi.org/10.1108/S0731-90532021000043A004
Polbin A., Shumilov A. (2025) Nowcasting and Forecasting Russian GDP and its Components Using Quantile Models. Applied Econometrics, 79, pp. 5–26. (In Russ.)
Rybak K.S. (2023) Evaluating the Role of Global Factors in GDP Nowcasting. Russian Economic Development, 3, 12, pp. 18–22. (In Russ.)
Stankevich I. (2023) Application of Markov-Switching MIDAS Models to Nowcasting of GDP and its Components. Applied Econometrics, 2, 70, pp. 122–143. (In Russ.) DOI: https://doi.org/10.22394/1993-7601-2023-70-122-143
Taylor J.W. (2000) A Quantile Regression Neural Network Approach to Estimating the Conditional Density of Multiperiod Returns. Journal of Forecasting, 19, 4, pp. 299–311. DOI: https://doi.org/10.1002/1099-131X(200007)19:4<299::AID-FOR775>3.3.CO;2-M
Zubarev A.V., Rybak K.S. (2021) GDP Nowcasting: Dynamic Factor Model vs. Official Forecasts. Russian Economic Development, 28, 12, pp. 34–40. (In Russ.)
Zubarev A., Lomonosov D., Rybak K. (2022) Estimation of the Impact of Global Shocks on the Russian Economy and GDP Nowcasting Using a Factor Model. Russian Journal of Money and Finance, 81, 2, pp. 49–78.
