Estimación adelantada del crecimiento regional mediante redes neuronales LSTM

  1. Juan de Lucio
Revista:
Investigaciones Regionales = Journal of Regional Research

ISSN: 1695-7253 2340-2717

Año de publicación: 2021

Número: 49

Páginas: 45-64

Tipo: Artículo

Otras publicaciones en: Investigaciones Regionales = Journal of Regional Research

Resumen

El trabajo propone incorporar técnicas de Inteligencia Artificial a las herramientas disponibles para el análisis de coyuntura regional. Se comparan las estimaciones realizadas con Redes Neuronales (en concreto, mediante la utilización de redes con larga memoria de corto plazo, LSTM por sus siglas en inglés) con los instrumentos más habituales en el análisis de coyuntura (series temporales, indicadores sintéticos y factores dinámicos). Los resultados muestran que los avances en redes neuronales pueden ser incorporados al análisis de coyuntura mejorando las estimaciones. Son herramientas complementarias, con mayor flexibilidad para captar la diversidad de situaciones en la economía real y con una capacidad de estimación superior (menor error cuadrático medio). El documento propone la utilización de este tipo de técnicas para solucionar una diversidad de problemas en economía regional.

Información de financiación

El autor agradece los comentarios de la editora y de dos evaluadores an?nimos y el apoyo financiero de la Comunidad de Madrid y la UAH (EPU-INV/2020/006) y de la Comunidad de Madrid (H2019/HUM-5761).

Financiadores

Referencias bibliográficas

  • Abadi, M., Agarwal, A., Barham, P., Brevdo, E., Chen, Z., Citro, C., Corrado, G. S., Davis, A., Dean, J., Devin, M., Ghemawat, S., Goodfellow, I., Harp, A., Irving, G., Isard, M., Jia, Y., Jozefowicz, R., Kaiser, L., Kudlur, M., Levenberg, J., Man_e, D., Monga, R., Moore, S., Murray, D., Olah, C., Schuster, M., Shlens, J., Steiner, B., Sutskever, I., Talwar, K., Tucker, P., Vanhoucke, V., Vasudevan, V., Vi_egas, F., Vinyals, O., Warden, P., Wattenberg, M., Wicke, M., Yu, Y., y Zheng, X. (2015). TensorFlow: Large-scale machine learning on heterogeneous systems. Software available from tensorow.org
  • Abadie, A., y Kasy, M. (2019). Choosing among Regularized Estimators in Empirical Economics: The Risk of Machine Learning. Review of Economics and Statistics Volume 101(5), 743-762.
  • Alcocer, U. M. R., Tello-Leal, E., y Alvarado, A. B. R. (2018). Modelo basado en redes neuronales LSTM para la predicción de la siguiente actividad en el proceso de negocio. Pistas Educativas, 40(130).
  • Artola, C., Fiorito, A., Gil, M., Pérez, J. J., Urtasun, A., y Vila, D. (2018). Monitoring the Spanish economy from a regional perspective: main elements of analysis. Documento Ocasional Banco de España 1809.
  • Bandrés, E., y Gadea, M. D., (2013). Crisis Económica y Ciclos Regionales en España. Papeles de Economía Española, 138, 2-30.
  • Bates, J. M., y Granger, C. W. J. (1969). The Combination of Forecasts. Operations Research Quarterly 20, 451–468.
  • Bengio, Y., Simard, P., y Frasconi, P. (1994). Learning long-term dependencies with gradient descent is difficult. IEEE Transactions on Neural Networks, 5(2), 157–166.
  • Box, G. E. P., y Jenkins, G. M. (1970). Time Series Analysis: Forecasting and Control. San Francisco.
  • Casares, F. F. (2017). Nowcasting: Modelos de Factores Dinámicos y Ecuaciones Puente para la Proyección del PIB del Ecuador. COMPENDIUM: Cuadernos de Economía y Administración, 4(8), 25-46.
  • Chatterjee, A., Ayadi, O., y Boone, B. (2000). Artificial Neural Network and the Financial Markets: A Survey. Managerial Finance, 26(12), 32-45.
  • Chollet, F. et al. (2015). Keras. https://github.com/keras-team/keras
  • Claveria, O., Monte, E., y Torra, S. (2018). A regional perspective on the accuracy of machine learning forecasts of tourism demand based on data characteristics. IREA Working Papers. University of Barcelona, Research Institute of Applied Economics.
  • Crone, S. F., Hibon, M., y Nikolopoulos, K. (2011). Advances in forecasting with neural networks? Empirical evidence from the NN3 competition on time series prediction. International Journal of Forecasting, 27(3), 635–660.
  • Cuevas, A., y Quilis, E. M. (2015). Quarterly Regional GDP Flash Estimates for the Spanish Economy (METCAP model). Autoridad Independiente de Responsabilidad Fiscal AIReF, Working Paper n. 3.
  • de Lucio, J., y Cardenete A. (2019). Predicción económica regional para la toma de decisiones. La Riqueza de las Regiones: Aportaciones de la Ciencia Regional a la Sociedad, 183-196. Thomson Reuters.
  • Gadea, M. D., Gómez-Loscos, A., y Montañés A. (2012). Cycles inside cycles: Spanish regional aggregation. SERIEs Journal of the Spanish Economic Association, 3, 423-456.
  • Giannone, D., Reichlin, L., y Small, D. (2008). Nowcasting: The real-time informational content. Journal of Monetary Economics 55(4), 665-676.
  • Gil, M., Leiva-León, D., Pérez, J. J., y Urtasun, A. (2018). An application of dynamic factor models to nowcast regional economic activity. Banco de España, mimeo.
  • Glaeser, E., Duke S., Luca M., y Naik N. (2018). Big Data and Big Cities: The Promises and Limitations of Improved Measures of Urban Life. Economic Inquiry, 56, 114-137. https://doi.org/10.1111/ecin.12364
  • Hewamalage, H., y Bergmeir, C. (2019). Recurrent Neural Networks for Time Series Forecasting: Current Status and Future Directions. Mimeo.
  • Hill, T., Marquez, L., O’Connor, M., y Remus, W. (1994). Artificial Neural Network Models for Forecasting and Decision Making. International Journal of Forecasting, 10, 5–15.
  • Hochreiter, S., y Schmidhuber, J. (1997). Long short-term memory. Neural Computation. 9(8), 1735– 1780. https://doi.org/10.1162/neco.1997.9.8.1735
  • Huang, W., Lai, K. K., Nakamori, Y., Wang, S., y Yu, L. (2007). Neural networks in finance and economics forecasting. International Journal of Information Technology & Decision Making, 6(01), 113-140.
  • Hunter, J. D. (2007). Matplotlib: A 2D graphics environment. Computing in Science & Engineering, 9, 90–95.
  • INE (1993). Contabilidad Nacional Trimestral de España (CNTR). Metodología y serie trimestral 1970- 1992. Instituto Nacional de Estadística.
  • Kingma, D. P., y Ba, J. (2014). Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980.
  • Kleinberg, J., Ludwig, J., Mullainathan, S., y Obermeyer, Z. (2015). Prediction policy problems. American Economic Review, 105(5), 491-95.
  • Lehmann, R., y Wohlrabe, K. (2014). Regional economic forecasting: state-of-the-art methodology and future challenges. Economic and Business Letters, 3(4), 218–231.
  • Longhi, S., Nijkamp, P., Reggiani, A., y Maierhofer, E. (2005). Neural network modeling as a tool for forecasting regional employment patterns. International Regional Science Review, 28(3), 330-346.
  • López, A. M. (2016). El papel de la información económica como generador de conocimiento en el proceso de predicción: comparaciones empíricas del crecimiento del PIB regional. Estudios de Economía Aplicada, 34, 553-582.
  • López, A. M. y Castro R. (2004). Valoración de la actividad económica regional de España a través de indicadores sintéticos. Estudios de Economía Aplicada, 22, 631-655.
  • Makridakis, S., Spiliotis, E., y Assimakopoulos, V. (2018). Statistical and Machine Learning forecasting methods: Concerns and ways forward. PLoS ONE 13(3): e0194889. https://doi.org/10.1371/journal.pone.0194889
  • McKinney, W. (2010). Data structures for statistical computing in Python. En S. van der Walt y J. Millman (Eds.), Proceedings of the 9th Python in science conference.
  • Moshiri, S., y Cameron, N. (2000). Neural network versus econometric models in forecasting inflation. Journal Forecast, 19, 201-217
  • Mullainathan, S., y Spiess, J. (2017). Machine Learning: An Applied Econometric Approach. Journal of Economic Perspectives, 31(2), 87-106.
  • Olah, C. (2015). Understanding LSTM Networks. http://colah.github.io/posts/2015-08-UnderstandingLSTMs/
  • Patuelli, R., Longhi, S., Nijkamp, P., y Reggiani, A. (2008). Neural networks and genetic algorithms as forecasting tools: a case study on German regions. Environment and Planning B: Planning and Design, 35(4), 701-722.
  • Patuelli, R., Longhi, S., Nijkamp, P., Reggiani, A., y Blien, U. (2007). A rank-order test on the statistical performance of neural network models for regional labor market forecasts. The Review of Regional Studies, 37(1), 64-81.
  • Pinkwart, N. (2018). Short-term forecasting economic activity in Germany: A supply and demand side system of bridge equations. Discussion Papers 36. Deutsche Bundesbank.
  • Ramajo, J., Márquez, M. A., y Hewings, G. J. D. (2015). Spatiotemporal Analysis of Regional Systems: A Multiregional Spatial Vector Autoregressive Model for Spain. International Regional Science Review, 40, 75-96.
  • Seabold, S., y Perktold, J. (2010). Statsmodels Econometric and statistical modeling with python. Proceedings of the 9th Python in Science Conference.
  • Siami-Namini, S., y Namin, A. S. (2018). Forecasting economics and financial time series: ARIMA vs. LSTM. arXiv preprint arXiv:1803.06386.
  • Stock, J.H., y Watson, M. (2001). A Comparison of Linear and Nonlinear Univariate Models for Forecasting Macroeconomic Time Series. En R.F. Engle y H. White (Eds.), Festschrift in Honour of Clive Granger (pp.1-44). Cambridge University Press.
  • Tkáč, M., y Verner, R. (2016). Artificial neural networks in business: Two decades of research. Applied Soft Computing, 38, 788-804.
  • Trujillo, F., Benítez, M. D., y López, P. (1999). Indicadores Sintéticos Trimestrales de la Actividad Económica No Agraria en Andalucía. Revista de Estudios Regionales, 53, 97-128.
  • Van Der Walt, S., Chris Colbert, S., y Varoquaux, G. (2011). The NumPy array: a structure for efficient numerical computation. Computing in Science and Engineering, Institute of Electrical and Electronics Engineers, 13(2), 22-30.
  • Varian, H. R. (2014). Big Data: New Tricks for Econometrics. Journal of Economic Perspectives, 28(2), 3–28.
  • Woloszko, N. (2018). Economic Modeling & Machine Learning, OECD. https://techpolicyinstitute.org/wp-content/uploads/2018/02/Woloszko_Macroeconomicforecasting-with-machine-learning-TPI.pdf
  • Zhang, G., Patuwo, B. E., y Hu, M. Y. (1998). Forecasting with artificial neural networks: The state of the art. International Journal of Forecasting, 14(1), 35–62.