The development of sophisticated forecasting and modeling techniques based on statistical learning or Machine Learning has led to the application of these algorithms in various fields, including finance. In this paper, we described two methodologies that combined statistical learning techniques, i.e., Support Vector Regression (SVR) and Local Regression (LR), with the recursive or bootstrapping method to obtain the term structure of the interest rates in the Spanish public debt market in 2020. The analysis was replicated for 2023, enabling the robustness of the competing methodologies to be evaluated under a markedly different monetary-policy environment. These estimates were then compared with those obtained by the Nelson-Siegel model by calculating the theoretical price of the bonds in the sample. Although the Nelson-Siegel model remained one of the most popular methods for estimating the term structure of interest rates, the methods based on machine learning algorithms also performed well. Thus, our findings suggested that, in the Spanish government bond market during 2020, the Nelson-Siegel model outperforms the machine-learning alternatives considered, although the latter provides flexible and economically plausible estimates of the term structure of interest rates. Nevertheless, SVR generated the lowest average pricing errors in 2023. Consequently, the results indicate that the effectiveness of yield-curve estimation methodologies depends on prevailing market conditions.
Citation: Marta Tolentino, Pedro Gento, Francisco Jareño, Juan Navas-Ureña. Alternative methods for estimating the term structure of interest rates: Some evidence from the Spanish market[J]. National Accounting Review, 2026, 8(4): 513-538. doi: 10.3934/NAR.2026021
The development of sophisticated forecasting and modeling techniques based on statistical learning or Machine Learning has led to the application of these algorithms in various fields, including finance. In this paper, we described two methodologies that combined statistical learning techniques, i.e., Support Vector Regression (SVR) and Local Regression (LR), with the recursive or bootstrapping method to obtain the term structure of the interest rates in the Spanish public debt market in 2020. The analysis was replicated for 2023, enabling the robustness of the competing methodologies to be evaluated under a markedly different monetary-policy environment. These estimates were then compared with those obtained by the Nelson-Siegel model by calculating the theoretical price of the bonds in the sample. Although the Nelson-Siegel model remained one of the most popular methods for estimating the term structure of interest rates, the methods based on machine learning algorithms also performed well. Thus, our findings suggested that, in the Spanish government bond market during 2020, the Nelson-Siegel model outperforms the machine-learning alternatives considered, although the latter provides flexible and economically plausible estimates of the term structure of interest rates. Nevertheless, SVR generated the lowest average pricing errors in 2023. Consequently, the results indicate that the effectiveness of yield-curve estimation methodologies depends on prevailing market conditions.
| [1] |
Adrian T, Crump RK, Moench E (2013) Pricing the term structure with linear regressions. J Financ Econ 110: 110–138. https://doi.org/10.1016/j.jfineco.2013.04.009 doi: 10.1016/j.jfineco.2013.04.009
|
| [2] |
Badics MC, Huszar ZR, Kotro BB (2023) The impact of crisis periods and monetary decisions of the Fed and the ECB on the sovereign yield curve network. J Int Financ Mark Inst Money 88: 101837. https://doi.org/10.1016/j.intfin.2023.101837 doi: 10.1016/j.intfin.2023.101837
|
| [3] | Bracke P, Datta A, Jung C, et al. (2019) Machine learning explainability in finance: an application to default risk analysis. Bank of England Staff Working Paper No. 816. https://doi.org/10.2139/ssrn.3435104 |
| [4] |
Buse A (1970) Expectations, prices, coupons and yields. J Financ 25: 809–818. https://doi.org/10.1111/j.1540-6261.1970.tb00555.x doi: 10.1111/j.1540-6261.1970.tb00555.x
|
| [5] |
Carleton WT, Cooper IA (1976) Estimation and uses of the term structure of interest rates. J Financ 31: 1067–1083. https://doi.org/10.1111/j.1540-6261.1976.tb01960.x doi: 10.1111/j.1540-6261.1976.tb01960.x
|
| [6] |
Christensen JHE, Diebold FX, Rudebusch GD (2011) The affine arbitrage-free class of Nelson–Siegel term structure models. J Econom 164: 4–20. https://doi.org/10.1016/j.jeconom.2011.02.011 doi: 10.1016/j.jeconom.2011.02.011
|
| [7] |
Colletaz G, Gourlaouen JP (1990) Cointégration et structure par terme des taux d'intérêt. Rev Econ 41: 687–712. https://doi.org/10.3406/reco.1990.409231 doi: 10.3406/reco.1990.409231
|
| [8] | Contreras D, Navarro E (1993) The use of exponential splines for the estimation of the intertemporal structure of the interest rates in the Spanish market. ASSET Meeting, Barcelona. |
| [9] | Cox JC, Ingersoll JE Jr, Ross SA (2005) A theory of the term structure of interest rates. In: Bhattacharya S, Constantinides GM (eds) Theory of Valuation, 2nd edn. Singapore: World Scientific, 129–164. https://doi.org/10.1142/9789812701022_0005 |
| [10] |
Díaz A, Jareño F, Navarro E (2011) Term structure of volatilities and yield curve estimation methodology. Quant Financ 11: 573–586. https://doi.org/10.1080/14697680903473286 doi: 10.1080/14697680903473286
|
| [11] |
Diebold FX, Li C (2006) Forecasting the term structure of government bond yields. J Econom 130: 337–364. https://doi.org/10.1016/j.jeconom.2005.03.005 doi: 10.1016/j.jeconom.2005.03.005
|
| [12] |
Duffie D, Kan R (1996) A yield-factor model of interest rates. Math Financ 6: 379–406. https://doi.org/10.1111/j.1467-9965.1996.tb00123.x doi: 10.1111/j.1467-9965.1996.tb00123.x
|
| [13] | Elosua P (2011) Introducción al entorno R. Bilbao: Servicio Editorial de la Universidad del País Vasco. |
| [14] |
Forradellas RR, Cabrera DS, Garay Gallastegui LM, et al. (2026) Characterization of S & P 500 companies by sector using artificial intelligence: statistical evidence and machine learning application. J Financ Data Sci 12: 100193. https://doi.org/10.1016/j.jfds.2026.100193 doi: 10.1016/j.jfds.2026.100193
|
| [15] | Goodfriend M (1998) Using the term structure of interest rates for monetary policy. Fed Reserve Bank Richmond Econ Q 84: 13–30. |
| [16] |
He J, Makarov RN, Tuero J, et al. (2024) Performance evaluation metric for statistical learning trading strategies. Data Sci Financ Econ 4: 570–600. https://doi.org/10.3934/DSFE.2024024 doi: 10.3934/DSFE.2024024
|
| [17] |
He X, Ng P (1999) COBS: qualitatively constrained smoothing via linear programming. Comput Stat 14: 315–337. https://doi.org/10.1007/s001800050019 doi: 10.1007/s001800050019
|
| [18] |
Heath D, Jarrow R, Morton A (1992) Bond pricing and the term structure of interest rates: a new methodology for contingent claims valuation. Econometrica 60: 77–105. https://doi.org/10.2307/2951677 doi: 10.2307/2951677
|
| [19] | James G, Witten D, Hastie T, et al. (2013) An Introduction to Statistical Learning: with Applications in R. New York: Springer. https://doi.org/10.1007/978-1-4614-7138-7 |
| [20] |
Laurini MP, Moura M (2010) Constrained smoothing B-splines for the term structure of interest rates. Insur Math Econ 46: 339–350. https://doi.org/10.1016/j.insmatheco.2009.11.008 doi: 10.1016/j.insmatheco.2009.11.008
|
| [21] |
Mestiri S (2024) Credit scoring using machine learning and deep learning-based models. Data Sci Financ Econ 4: 236–248. https://doi.org/10.3934/DSFE.2024009 doi: 10.3934/DSFE.2024009
|
| [22] |
Mineo E, Alencar AP, Moura M, et al. (2020) Forecasting the term structure of interest rates with dynamic constrained smoothing B-splines. J Risk Financ Manag 13: 65. https://doi.org/10.3390/jrfm13040065 doi: 10.3390/jrfm13040065
|
| [23] | Nave-Pineda JM, Contreras-Bayarri D, Ferrer-Lapeña R, et al. (1996) Análisis factorial de la estructura temporal de los tipos de interés en España. Rev Esp Financ Contab 25: 139–160. |
| [24] |
Nelson CR, Siegel AF (1987) Parsimonious modeling of yield curves. J Bus 60: 473–489. https://doi.org/10.1086/296409 doi: 10.1086/296409
|
| [25] | Núñez Ramos S (1995) Estimación de la estructura temporal de los tipos de interés en España: elección entre métodos alternativos. Documentos de Trabajo No. 9522. Madrid: Banco de España. |
| [26] | Robles-Fernández MD, Abad-Romero P (2003) Estructura temporal de los tipos de interés: teoría y evidencia empírica. Rev Asturiana Econ 27: 7–47. |
| [27] | Svensson LEO (1994) Estimating and interpreting forward interest rates: Sweden 1992–1994. NBER Working Paper No. 4871. https://doi.org/10.3386/w4871 |
| [28] |
Umar Z, Riaz Y, Aharon DY (2022) Network connectedness dynamics of the yield curve of G7 countries. Int Rev Econ Financ 79: 275–288. https://doi.org/10.1016/j.iref.2022.02.052 doi: 10.1016/j.iref.2022.02.052
|
| [29] |
Vasicek O (1977) An equilibrium characterization of the term structure. J Financ Econ 5: 177–188. https://doi.org/10.1016/0304-405X(77)90016-2 doi: 10.1016/0304-405X(77)90016-2
|
| [30] |
Vasicek OA, Fong HG (1982) Term structure modeling using exponential splines. J Financ 37: 339–348. https://doi.org/10.1111/j.1540-6261.1982.tb03555.x doi: 10.1111/j.1540-6261.1982.tb03555.x
|
| [31] | Yang H, Chan L, King I (2002) Support vector machine regression for volatile stock market prediction. In: Intelligent Data Engineering and Automated Learning — IDEAL 2002. Lecture Notes in Computer Science, vol 2412. Berlin, Heidelberg: Springer, 391–396. https://doi.org/10.1007/3-540-45675-9_58 |