Ítem
Acceso Abierto
Forecasting Dynamic Term Structure Models with Autoencoders
| dc.contributor.gruplac | Grupo de investigaciones. Facultad de Economía. Universidad del Rosario | spa |
| dc.creator | Castro, Carlos | |
| dc.creator | Ramirez, Julian | |
| dc.date.accessioned | 2021-07-30T17:19:38Z | |
| dc.date.available | 2021-07-30T17:19:38Z | |
| dc.date.created | 2021-05-10 | |
| dc.date.issued | 2021-07-30 | |
| dc.description | El Análisis de Componentes Principales (PCA) es un método estadístico para construir modelos factoriales en finanzas. PCA es también un caso particular de un tipo de red neuronal conocido como Autoencoder. Recientemente los autoencoders han sido utilizados satisfactoriamente en modelos factoriales en finanzas, Gu et al. (2020), Heaton y Polson (2017). En este documento estudiamos la relación entre autoencoders y modelos de estructura a plazos de la tasa de interés dinámicos; adicionalmente proponemos varias alternativas para generar pronósticos. Comparamos el desempeño de pronóstico de los modelos factoriales basados en autoencoders, modelos clásicos de la estructura a plazos de la tasa de interés propuestos en Diebold y Li (2006) y modelos de redes neuronales para series de tiempo. Nuestro ejercicio empírico evalúa el desempeño de pronóstico de los autoencoders con información de la curva de tasa de interés de los tesoros del gobierno norteamericano durante ellos últimos 35 años. Los resultados preliminares indican que un modelo hibrido de autoencoders y vectores autorregresivos, representado como un modelo dinámico de la estructura a plazos de la tasa de interés, proporciona un pronóstico adecuado y consistente a lo largo de la muestra. Este modelo hibrido supera los inconvenientes de sobre-ajuste dentro de la muestra de otros modelos y los cambios estructurales observados en los datos. | spa |
| dc.description.abstract | Principal Components Analysis (PCA) is a statistical approach to build factor models in finance. PCA is also a particular case of a type of neural network known as an autoencoder. Recently, autoencoders have been successfully applied in financial applications using factor models, Gu et al. (2020), Heaton and Polson (2017). We study the relationship between autoencoders and dynamic term structure models; furthermore we propose different approaches for forecasting. We compare the forecasting accuracy of dynamic factor models based on autoencoders, classical models in term structure modelling proposed in Diebold and Li (2006) and neural network-based approaches for time series forecasting. Empirically, we test the forecasting performance of autoencoders using the U.S. yield curve data in the last 35 years. Preliminary results indicate that a hybrid approach using autoencoders and vector autoregressions framed as a dynamic term structure model provides an accurate forecast that is consistent throughout the sample. This hybrid approach overcomes in-sample overfitting and structural changes in the data. | spa |
| dc.format.extent | 36 pp. | spa |
| dc.format.mimetype | application/pdf | |
| dc.identifier.doi | https://doi.org/10.48713/10336_31955 | |
| dc.identifier.uri | https://repository.urosario.edu.co/handle/10336/31955 | |
| dc.language.iso | eng | spa |
| dc.publisher | Universidad del Rosario | |
| dc.publisher.department | Facultad de Economía | |
| dc.relation.ispartof | Documentos de trabajo economía, (2021); 36 pp. | spa |
| dc.relation.uri | https://ideas.repec.org/p/col/000092/019431.html | |
| dc.rights | Atribución-CompartirIgual 2.5 Colombia | * |
| dc.rights.accesRights | info:eu-repo/semantics/openAccess | |
| dc.rights.acceso | Abierto (Texto Completo) | spa |
| dc.rights.uri | http://creativecommons.org/licenses/by-sa/2.5/co/ | |
| dc.source.bibliographicCitation | Shihao Gu, Bryan T. Kelly, and Dacheng Xiu. Autoencoder asset pricing models. Journal of Econometrics, 2020. ISSN 1556-5068. doi: doi.org/ 10.1016/j.jeconom.2020.07.009. | spa |
| dc.source.bibliographicCitation | J. B. Heaton and Nick Polson. Deep learning for finance: Deep portfolios. Appl. Stoch. Models Bus. Ind., 2017. ISSN 1556-5068. doi: doi.org/10.1002/asmb.2209 | spa |
| dc.source.bibliographicCitation | Francis Diebold and Canlin Li. Forecasting the term structure of government bond yields. J. Econometrics, 2006. doi: 10.1016/j.jeconom. 2005.03.005 | spa |
| dc.source.bibliographicCitation | Robert Engle, Guillaume Roussellet, and Emil Siriwardane. Scenario generation for long run interest rate risk assessment. 201:333–347, 2017. ISSN 0304-4076. doi: 10.1016/j.jeconom.2017.08.012 | spa |
| dc.source.bibliographicCitation | Charles R. Nelson and Andrew F. Siegel. Parsimonious modeling of yield curves. Journal of Business, 60:473, 1987. ISSN 0021-9398. doi: 10.1086/296409. | spa |
| dc.source.bibliographicCitation | Jens H. E. Christensen, Francis X. Diebold, and Glenn D. Rudebusch. The affine arbitrage-free class of nelsonsiegel term structure models. Journal of Econometrics, 164:4–20, 2011. ISSN 0304-4076. doi: 10.1016/j.jeconom.2011.02.011 | spa |
| dc.source.bibliographicCitation | D. Redfern and D. McLean. Principal component analysis for yield curve modelling: Reproduction of out-of-sample yield curves. Technical report, Moodys Analytics, 2014. | spa |
| dc.source.bibliographicCitation | Marcos L´opez de Prado. Advances in Financial Machine Learning:. Wiley, 2018. doi: 10.2139/ssrn.3270269 | spa |
| dc.source.bibliographicCitation | Matthew F. Dixon, Igor Halperin, and Paul Bilokon. Machine Learning in Finance: From theory to practice. Springer, 2020. doi: 10.1007/ 978-3-030-41068-1 | spa |
| dc.source.bibliographicCitation | Hansika Hewamalage, Christoph Bergmeir, and Kasun Bandara. Recurrent neural networks for time series forecasting: Current status and future directions. International Journal of Forecasting, 37:388–427, 2021. ISSN 0169-2070. doi: 10.1016/j.ijforecast.2020.06.008 | spa |
| dc.source.bibliographicCitation | Anastasia Borovykh, Sander Bohte, and Cornelis W. Oosterlee. Conditional time series forecasting with convolutional neural networks. March 2017 | spa |
| dc.source.bibliographicCitation | G. P. Zhang and V. L. Berardi. Time series forecasting with neural network ensembles: an application for exchange rate prediction. Journal of the Operational Research Society, 52:652–664, 2001. ISSN 0160-5682. doi: 10.1057/palgrave.jors.2601133. | spa |
| dc.source.bibliographicCitation | Yoshiyuki Suimon, Hiroki Sakaji, Kiyoshi Izumi, and Hiroyasu Matsushima. Autoencoder-based three-factor model for the yield curve of japanese government bonds and a trading strategy. 13:82, 2020. ISSN 1911-8074. doi: 10.3390/jrfm13040082 | spa |
| dc.source.bibliographicCitation | G. Kirczenow, A. Fathi, and M. Davison. Machine learning for yield curve feature extraction: Application to illiquid corporate bonds. June 2018. | spa |
| dc.source.bibliographicCitation | Alexei Kondratyev. Learning curve dynamics with artificial neural networks. April 2018 | spa |
| dc.source.bibliographicCitation | P. Baldi and K. Hornik. Neural networks and principal component analysis: Learning from examples without local minima. Neural Networks, 2:53–58, 1989 | spa |
| dc.source.bibliographicCitation | Lars E. O. Svensson. Estimating and interpreting forward interest rates: Sweden 1992 - 1994. NBER Working Paper Series, 1994. 4871 | spa |
| dc.source.bibliographicCitation | Michiel De Pooter. Examining the nelson-siegel class of term structure models: In-sample fit versus out-of-sample forecasting performance. 2007. ISSN 1556-5068. doi: 10.2139/ssrn.992748 | spa |
| dc.source.bibliographicCitation | R.J. Hyndman, Koehler A.B., J.K. Ord, and R.D. Snyder. Forescasting with exponential smoothing: The state space approach. Springer, 2008 | spa |
| dc.source.bibliographicCitation | Martin M. Andreasen, Jens H. E. Christensen, and Glenn D. Rudebusch. Term structure analysis with big data: One-step estimation using bond prices. Journal of Econometrics, 212:26–46, 2019. ISSN 0304-4076. doi: 10.1016/j.jeconom.2019.04.019 | spa |
| dc.source.bibliographicCitation | Hitesh Doshi, Kris Jacobs, and Rui Liu. Information in the term structure: A forecasting perspective. Management Science, 2020. ISSN 0025- 1909. doi: 10.1287/mnsc.2020.3715 | spa |
| dc.source.bibliographicCitation | Francis Diebold, Glenn Rudebusch, and S. Boragan Aruoba. The macroeconomy and the yield curve: A dynamic latent factor approach. Journal of Econometrics, (131):309–338, 2006. doi: 10.3386/w10616. | spa |
| dc.source.bibliographicCitation | Andrew Ang and Monika Piazzesi. A no-arbitrage vector autoregression of term structure dynamics with macroeconomic and latent variables. Journal of Monetary Economics, 50:745–787, 2003. ISSN 0304-3932. doi: 10.1016/s0304-3932(03)00032-1 | spa |
| dc.source.bibliographicCitation | Giang Nguyen, Robert Engle, Michael Fleming, and Eric Ghysels. Liquidity and volatility in the u.s. treasury market. Journal of Econometrics, 217:207–229, 2020. ISSN 0304-4076. doi: 10.1016/j.jeconom.2019. 12.002 | spa |
| dc.source.bibliographicCitation | Mardi Dungey, Michael McKenzie, and L. Vanessa Smith. Empirical evidence on jumps in the term structure of the us treasury market. Journal of Empirical Finance, 16:430–445, 2009. ISSN 0927-5398. doi: 10.1016/j.jempfin.2008.12.002. | spa |
| dc.source.bibliographicCitation | David Harvey, Stephen Leybourne, and Paul Newbold. Testing the equality of prediction mean squared errors. International Journal of Forecasting, 13:281–291, 1997. ISSN 0169-2070. doi: 10.1016/ s0169-2070(96)00719-4 | spa |
| dc.source.bibliographicCitation | F. Van Veen and S. Leijnen. The neural network zoo, 2019. URL https://www.asimovinstitute.org/neural-network-zoo | spa |
| dc.source.instname | instname:Universidad del Rosario | |
| dc.source.reponame | reponame:Repositorio Institucional EdocUR | |
| dc.subject | Uso de redes neuronales recurrentes para la estimación de tasas de interés | spa |
| dc.subject | Comparación entre Autoencoder de modelos de estructura a plazos en la determinación de tasas de interés dinámicas | spa |
| dc.subject | Uso de redes neuronales recurrentes en evaluación financiera | spa |
| dc.subject | Modelos factoriales y estructura a plazos de tasas de interés | spa |
| dc.subject | Comparación entre de técnicas predictivas de tasas de interés en finanzas | spa |
| dc.subject.ddc | Economía financiera | spa |
| dc.subject.jel | C45 | spa |
| dc.subject.jel | C53 | spa |
| dc.subject.jel | C58 | spa |
| dc.subject.keyword | Comparison of predictive techniques of interest rates in surety bonds | spa |
| dc.subject.keyword | Use of recurrent neural networks to estimate interest rates | spa |
| dc.subject.keyword | Comparison between Autoencoder of term structure models in the determination of dynamic interest rates | spa |
| dc.subject.keyword | Use of recurrent neural networks in financial evaluation | spa |
| dc.subject.keyword | Factor models and term structure of interest rates | spa |
| dc.title | Forecasting Dynamic Term Structure Models with Autoencoders | spa |
| dc.title.TranslatedTitle | Pronóstico de modelos de estructura temporal dinámica con Autocodificador | spa |
| dc.type | workingPaper | eng |
| dc.type.hasVersion | info:eu-repo/semantics/draft | |
| dc.type.spa | Documento de trabajo | spa |
