voprecotestВопросы экономикиVoprosy Ekonomiki0042-8736Voprosy Ekonomiki, NP10.32609/0042-8736-2022-8-133-157voprecotest-4093Research ArticleМЕТОДОЛОГИЯ ЭКОНОМИЧЕСКОГО АНАЛИЗАMETHODOLOGY OF ECONOMIC ANALYSISНаукастинг: оценка изменения ключевых макроэкономических показателей с использованием методов машинного обученияNowcasting Russia’s key macroeconomic variables using machine learninghttps://orcid.org/0000-0002-7208-291XГареевМ. Ю.GareevM. Y.

Гареев Михаил Юрьевич, м. н. с. лаборатории математического моделирования экономических процессов Института прикладных экономических исследований

Москва

Mikhail Y. Gareev

Moscow

 

mkhlgrv@gmail.comhttps://orcid.org/0000-0003-4683-8194ПолбинА. В.PolbinA. V.

Полбин Андрей Владимирович, к. э. н., завлабораторией математического моделирования экономических процессов;, зам. зав. международной лабораторией математического моделирования экономических процессов 

Москва

Andrey V. Polbin

Moscow

apolbin@iep.ruРоссийская академия народного хозяйства и государственной службы при Президенте РФРоссияRussian Presidential Academy of National Economy and Public AdministrationRussian FederationРоссийская академия народного хозяйства и государственной службы при Президенте РФ; Институт экономической политики имени Е. Т. ГайдараРоссияRussian Presidential Academy of National Economy and Public Administration; Gaidar Institute for Economic PolicyRussian Federation20220408202208133—157Copyright © Voprosy Ekonomiki, NP, 20222022Voprosy Ekonomiki, NPVoprosy Ekonomiki, NPhttps://www.vopreco.ru/jour/about/submissions#copyrightNoticehttps://www.vopreco.ru/jour/article/view/4093

Разработана методика наукастинга и краткосрочного прогнозирования квартальных изменений ключевых макроэкономических показателей — ВВП, потребления, инвестиций, показателей внешней торговли — с помощью методов машинного обучения: бустинга, эластичной сети и случайного леса. В рамках эксперимента в качестве предикторов использовались показатели фондов ого и денежного рынков, опросов, мировые цены на ресурсы, индексы цен и другие статистические показатели разной периодичности. Такой подход позволил детально рассмотреть изменение прогнозов по мере поступления новой информации в течение квартала. Для большинства показателей получено монотонное неухудшение качества прогнозов с ростом доступной информации. Методы машинного обучения продемонстрировали значительное превосходство в качестве предсказания по сравнению с наивным прогнозом. Рассмотренные методы в рамках псевдоэксперимента уже после десятой недели квартала идентифицировали сильное падение ВВП, потребления и других показателей в условиях развития пандемии COVID-19 во II и III кв. 2020 г.

The article developed a methodology for nowcasting and short-term forecasting key Russian macroeconomic aggregates: real GDP, consumption, investment, export, import, using machine learning methods: boosting, elastic net, and random forest. The set of predictors included indicators of the stock market, money market, surveys, world prices for resources, price indices, and other statistical indicators of different frequency, from daily to quarterly. Our approach makes available a detailed examination of the changes in forecasts with the flow of new information. For most of the considered variables, a monotonic non-deterioration of the forecast quality was obtained with an expansion of available information. Furthermore, machine learning methods have shown significant superiority in predictive performance over naive prediction. The considered methods within the framework of the pseudo-experiment quickly showed a strong drop in real GDP, household consumption, and other variables in the context of the spread of the COVID-19 pandemic in the 2nd and 3rd quarters of 2020.

макроэкономикамакроэкономическое прогнозированиенаукастингмашинное обучениебустингслучайный лесбольшие данныеРоссияmacroeconomicsmacroeconomic forecastingnowcastingmachine learningboostingelastic netrandom forestbig dataRussiaReferencesГареев М. Ю. (2020). Использование методов машинного обучения для прогнозирования инвестиций в России // Деньги и кредит. Т. 79, № 1. С. 35—56. https://doi.org/10.31477/rjmf.202001.35Gareev M. Y. (2020). Use of machine learning methods to forecast investment in Russia. Russian Journal of Money and Finance, Vol. 79, No. 1, pp. 35—56. (In Russian). https://doi.org/10.31477/rjmf.202001.35Демешев Б. Б., Малаховская О. А. (2016). Макроэкономическое прогнозирование с помощью BVAR Литтермана // Экономический журнал Высшей школы экономики, Т. 20, № 4. С. 691—710.Demeshev B. B., Malakhovskaya O. A. (2016). Macroeconomic forecasting with a Litterman’s BVAR model. HSE Economic Journal, Vol. 20, No. 4, pp. 691—710. (In Russian).Пестова А. А., Мамонов М. Е. (2016). Оценка влияния различных шоков на динамику макроэкономических показателей в России и разработка условных прогнозов на основе BVAR-модели российской экономики // Экономическая политика. Т. 11, № 4. С. 56—92. https://doi.org/10.18288/1994-51242016-4-03Pestova A. A., Mamonov M. E. (2016). Estimating the influence of different shocks on macroeconomic indicators and developing conditional forecasts on the basis of BVAR model for the Russian economy. Ekonomicheskaya Politika, Vol. 11, No. 4, pp. 56—92. (In Russian). https://doi.org/10.18288/1994-51242016-4-03Поршаков А. С., Пономаренко А. А., Синяков А. А. (2016). Оценка и прогнозирование ВВП России с помощью динамической факторной модели // Журнал Новой экономической ассоциации. Т. 30, № 2. С. 60—76. https://doi.org/10.31737/2221-2264-2016-30-2-3Porshakov A. S., Ponomarenko A. A., Sinyakov A. S. (2016). Nowcasting and short-term forecasting of Russian GDP with a dynamic factor model. Journal of the New Economic Association, Vol. 30, No. 2, pp. 60—76. (In Russian). https://doi.org/10.31737/2221-2264-2016-30-2-3Фокин Н. Д., Полбин А. В. (2019). VAR-LASSO модель для прогнозирования ключевых макроэкономических показателей России // Деньги и кредит. Т. 78, № 2. С. 67—93. https://doi.org/10.31477/rjmf.201902.67Fokin N. D., Polbin A. V. (2019). Forecasting Russia’s key macroeconomic indicators with the VAR-LASSO model. Russian Journal of Money and Finance, Vol. 78, No. 2, pp. 67—93. (In Russian). https://doi.org/10.31477/rjmf.201902.67Altman N. S. (1992). An introduction to kernel and nearest-neighbor nonparametric regression. The American Statistician, Vol. 46, No. 3, pp. 175—185. https:// doi.org/10.1080/00031305.1992.10475879Altman N. S. (1992). An introduction to kernel and nearest-neighbor nonparametric regression. The American Statistician, Vol. 46, No. 3, pp. 175—185. https:// doi.org/10.1080/00031305.1992.10475879Andreou E., Ghysels E., Kourtellos A. (2013). Should macroeconomic forecasters use daily financial data and how? Journal of Business & Economic Statistics, Vol. 31, No. 2, pp. 240—251. https://doi.org/10.1080/07350015.2013.767199Andreou E., Ghysels E., Kourtellos A. (2013). Should macroeconomic forecasters use daily financial data and how? Journal of Business & Economic Statistics, Vol. 31, No. 2, pp. 240—251. https://doi.org/10.1080/07350015.2013.767199Babii A., Ghysels E., Striaukas J. (2021). Machine learning time series regressions with an application to nowcasting. Journal of Business & Economic Statistics, Vol. 40, No. 3, pp. 1094—1106. https://doi.org/10.1080/07350015.2021.1899933Babii A., Ghysels E., Striaukas J. (2021). Machine learning time series regressions with an application to nowcasting. Journal of Business & Economic Statistics, Vol. 40, No. 3, pp. 1094—1106. https://doi.org/10.1080/07350015.2021.1899933Baffigi A., Golinelli R., Parigi G. (2004). Bridge models to forecast the euro area GDP. International Journal of Forecasting, Vol. 20, No. 3, pp. 447—460. https://doi.org/10.1016/S0169-2070(03)00067-0Baffigi A., Golinelli R., Parigi G. (2004). Bridge models to forecast the euro area GDP. International Journal of Forecasting, Vol. 20, No. 3, pp. 447—460. https://doi.org/10.1016/S0169-2070(03)00067-0Bańbura M., Giannone D., Modugno M., Reichlin L. (2013). Now-casting and the realtime data flow. In: G. Elliott (ed.). Handbook of economic forecasting, Vol. 2. Elsevier, pp. 195—237. https://doi.org/10.1016/B978-0-444-53683-9.00004-9Bańbura M., Giannone D., Modugno M., Reichlin L. (2013). Now-casting and the realtime data flow. In: G. Elliott (ed.). Handbook of economic forecasting, Vol. 2. Elsevier, pp. 195—237. https://doi.org/10.1016/B978-0-444-53683-9.00004-9Boser B. E., Guyon I. M., Vapnik V. N. (1992). A training algorithm for optimal margin classifiers. In: COLT’92: Proceedings of the Fifth annual workshop on computational learning theory. New York: Association for Computing Machinery, pp. 144—152. https://doi.org/10.1145/130385.130401Boser B. E., Guyon I. M., Vapnik V. N. (1992). A training algorithm for optimal margin classifiers. In: COLT’92: Proceedings of the Fifth annual workshop on computational learning theory. New York: Association for Computing Machinery, pp. 144—152. https://doi.org/10.1145/130385.130401Brave S. A., Butters R. A., Justiniano A. (2019). Forecasting economic activity with mixed frequency BVARs. International Journal of Forecasting, Vol. 35, No. 4, pp. 1692—1707. https://doi.org/10.1016/j.ijforecast.2019.02.010Brave S. A., Butters R. A., Justiniano A. (2019). Forecasting economic activity with mixed frequency BVARs. International Journal of Forecasting, Vol. 35, No. 4, pp. 1692—1707. https://doi.org/10.1016/j.ijforecast.2019.02.010Breiman L. (1994). Bagging predictors. Machine Learning, Vol. 24, No. 2, pp. 123—140. https://doi.org/10.1007/BF00058655Breiman L. (1994). Bagging predictors. Machine Learning, Vol. 24, No. 2, pp. 123—140. https://doi.org/10.1007/BF00058655Breiman L. (2001). Random forests. Machine Learning, Vol. 45, No. 1, pp. 5—32. https://doi.org/10.1023/A:1010933404324Breiman L. (2001). Random forests. Machine Learning, Vol. 45, No. 1, pp. 5—32. https://doi.org/10.1023/A:1010933404324Clements M. P., Galvão A. B. (2009). Forecasting US output growth using leading indicators: An appraisal using MIDAS models. Journal of Applied Econometrics, Vol. 24, No. 7, pp. 1187—1206. https://doi.org/10.1002/jae.1075Clements M. P., Galvão A. B. (2009). Forecasting US output growth using leading indicators: An appraisal using MIDAS models. Journal of Applied Econometrics, Vol. 24, No. 7, pp. 1187—1206. https://doi.org/10.1002/jae.1075Diebold F. X., Mariano R. S. (1995). Comparing predictive accuracy. Journal of Business and Economic Statistics, Vol. 13, No. 3, pp. 253—263. https://doi.org/10.1198/073500102753410444Diebold F. X., Mariano R. S. (1995). Comparing predictive accuracy. Journal of Business and Economic Statistics, Vol. 13, No. 3, pp. 253—263. https://doi.org/10.1198/073500102753410444Diron M. (2008). Short-term forecasts of euro area real GDP growth: An assessment of real-time performance based on vintage data. Journal of Forecasting, Vol. 27, No. 5, pp. 371—390. https://doi.org/10.1002/for.1067Diron M. (2008). Short-term forecasts of euro area real GDP growth: An assessment of real-time performance based on vintage data. Journal of Forecasting, Vol. 27, No. 5, pp. 371—390. https://doi.org/10.1002/for.1067Doz C., Giannone D., Reichlin L. (2011). A two-step estimator for large approximate dynamic factor models based on Kalman filtering. Journal of Econometrics, Vol. 164, No. 1, pp. 188—205. https://doi.org/10.1016/j.jeconom.2011.02.012Doz C., Giannone D., Reichlin L. (2011). A two-step estimator for large approximate dynamic factor models based on Kalman filtering. Journal of Econometrics, Vol. 164, No. 1, pp. 188—205. https://doi.org/10.1016/j.jeconom.2011.02.012Doz C., Giannone D., Reichlin L. (2012). A quasi-maximum likelihood approach for large, approximate dynamic factor models. Review of Economics and Statistics, Vol. 94, No. 4, pp. 1014—1024. https://doi.org/10.1162/REST_a_00225Doz C., Giannone D., Reichlin L. (2012). A quasi-maximum likelihood approach for large, approximate dynamic factor models. Review of Economics and Statistics, Vol. 94, No. 4, pp. 1014—1024. https://doi.org/10.1162/REST_a_00225Foroni C., Marcellino M. (2014). A comparison of mixed frequency approaches for nowcasting Euro area macroeconomic aggregates. International Journal of Forecasting, Vol. 30, No. 3, pp. 554—568. https://doi.org/10.1016/j.ijforecast.2013.01.010Foroni C., Marcellino M. (2014). A comparison of mixed frequency approaches for nowcasting Euro area macroeconomic aggregates. International Journal of Forecasting, Vol. 30, No. 3, pp. 554—568. https://doi.org/10.1016/j.ijforecast.2013.01.010Friedman J. H. (2001). Greedy function approximation: A gradient boosting machine. Annals of Statistics, Vol. 29, No. 5, pp. 1189—1232. https://doi.org/10.1214/aos/1013203451Friedman J. H. (2001). Greedy function approximation: A gradient boosting machine. Annals of Statistics, Vol. 29, No. 5, pp. 1189—1232. https://doi.org/10.1214/aos/1013203451Giannone D., Reichlin L., Small D. (2008). Nowcasting: The real-time informational content of macroeconomic data. Journal of Monetary Economics, Vol. 55, No. 4, pp. 665—676. https://doi.org/10.1016/j.jmoneco.2008.05.010Giannone D., Reichlin L., Small D. (2008). Nowcasting: The real-time informational content of macroeconomic data. Journal of Monetary Economics, Vol. 55, No. 4, pp. 665—676. https://doi.org/10.1016/j.jmoneco.2008.05.010Golinelli R., Parigi G. (2007). The use of monthly indicators to forecast quarterly GDP in the short run: An application to the G7 countries. Journal of Forecasting, Vol. 26, No. 2, pp. 77—94. https://doi.org/10.1002/for.1007Golinelli R., Parigi G. (2007). The use of monthly indicators to forecast quarterly GDP in the short run: An application to the G7 countries. Journal of Forecasting, Vol. 26, No. 2, pp. 77—94. https://doi.org/10.1002/for.1007Harvey D., Leybourne S., Newbold P. (1997). Testing the equality of prediction mean squared errors. International Journal of Forecasting, Vol. 13, No. 2, pp. 281—291. https://doi.org/10.1016/S0169-2070(96)00719-4Harvey D., Leybourne S., Newbold P. (1997). Testing the equality of prediction mean squared errors. International Journal of Forecasting, Vol. 13, No. 2, pp. 281—291. https://doi.org/10.1016/S0169-2070(96)00719-4Huber F., Koop G., Onorante L., Pfarrhofer M., Schreiner J. (2020). Nowcasting in a pandemic using non-parametric mixed frequency VARs. Journal of Econometrics, [forthcoming]. https://doi.org/10.1016/j.jeconom.2020.11.006Huber F., Koop G., Onorante L., Pfarrhofer M., Schreiner J. (2020). Nowcasting in a pandemic using non-parametric mixed frequency VARs. Journal of Econometrics, [forthcoming]. https://doi.org/10.1016/j.jeconom.2020.11.006Kuzin V., Marcellino M., Schumacher C. (2011). MIDAS vs. mixed-frequency VAR: Nowcasting GDP in the euro area. International Journal of Forecasting, Vol. 27, No. 2, pp. 529—542. https://doi.org/10.1016/j.ijforecast.2010.02.006Kuzin V., Marcellino M., Schumacher C. (2011). MIDAS vs. mixed-frequency VAR: Nowcasting GDP in the euro area. International Journal of Forecasting, Vol. 27, No. 2, pp. 529—542. https://doi.org/10.1016/j.ijforecast.2010.02.006Mikosch H., Solanko L. (2019). Forecasting quarterly Russian GDP growth with mixed frequency data. Russian Journal of Money and Finance, Vol. 78, No. 1, pp. 19—35. https://doi.org/10.31477/rjmf.201901.19Mikosch H., Solanko L. (2019). Forecasting quarterly Russian GDP growth with mixed frequency data. Russian Journal of Money and Finance, Vol. 78, No. 1, pp. 19—35. https://doi.org/10.31477/rjmf.201901.19Patton A. J., Timmermann A. (2012). Forecast rationality tests based on multi-horizon bounds. Journal of Business & Economic Statistics, Vol. 30, No. 1, pp. 1—17. https://doi.org/10.1080/07350015.2012.634337Patton A. J., Timmermann A. (2012). Forecast rationality tests based on multi-horizon bounds. Journal of Business & Economic Statistics, Vol. 30, No. 1, pp. 1—17. https://doi.org/10.1080/07350015.2012.634337Schorfheide F., Song D. (2015). Real-time forecasting with a mixed-frequency VAR. Journal of Business & Economic Statistics, Vol. 33, No. 3, pp. 366—380. https://doi.org/10.1080/07350015.2014.954707Schorfheide F., Song D. (2015). Real-time forecasting with a mixed-frequency VAR. Journal of Business & Economic Statistics, Vol. 33, No. 3, pp. 366—380. https://doi.org/10.1080/07350015.2014.954707Zhemkov M. (2021). Nowcasting Russian GDP using forecast combination approach. International Economics, Vol. 168, pp. 10—24. https://doi.org/10.1016/j.inteco.2021.07.006Zhemkov M. (2021). Nowcasting Russian GDP using forecast combination approach. International Economics, Vol. 168, pp. 10—24. https://doi.org/10.1016/j.inteco.2021.07.006Zou H., Hastie T. (2005). Regularization and variable selection via the elastic net. Journal of the Royal Statistical Society: Series B (Statistical Methodology), Vol. 67, No. 2, pp. 301—320. https://doi.org/10.1111/j.1467-9868.2005.00503.xZou H., Hastie T. (2005). Regularization and variable selection via the elastic net. Journal of the Royal Statistical Society: Series B (Statistical Methodology), Vol. 67, No. 2, pp. 301—320. https://doi.org/10.1111/j.1467-9868.2005.00503.x

The authors declare that there are no conflicts of interest present.