Early warning signals in the Brazilian yield curve
Published:
Jul 29, 2026
Volume:
24
Keywords:
Yield curve
Early warning system
LSTM
ARIMA
Abstract
This paper develops an early warning system (EWS) for changes in the Brazilian fixed-rate yield curve using ANBIMA coefficients from January 2015 to September 2025. We reconstruct the curve at nine maturities, convert the data to weekly averages, and estimate forecasting models in an expanding-window design that mimics real-time information. The main specification uses a multivariate LSTM with horizons of up to four weeks and compares its signals with an AutoARIMA benchmark under the same operational rules. We propose an ex ante alert procedure based on the median of recent weekly changes, dynamic maturity-specific thresholds, and signal aggregation across maturities. The LSTM system detects fewer isolated alerts and more persistent episodes than ARIMA. Alert weeks display higher cross-maturity volatility and a steeper yield curve. Although ARIMA achieves lower average RMSE and MAE, LSTM produces stronger regime contrast and more persistent alert dynamics.
How to cite
Guilherme C. C. Moreira, Daniel Oliveira Cajueiro, Douglas Silveira. Early warning signals in the Brazilian yield curve. Brazilian Review of Finance, v. 24, n. 1, 2026. p. e202612. DOI: 10.12660/rbfin.v24n1.2026.98237.
References
Arantes, B. D. O., Portugal, M. S. & Caldeira, J. F. (2014). Previsão da estrutura a termo da taxa de juros brasileira usando redes neurais artificiais, Anais do XLI Encontro Nacional de Economia [Proceedings of the 41st Brazilian Economics Meeting] 128, ANPEC - Associação Nacional dos Centros de Pós-Graduação em Economia [Brazilian Association of Graduate Programs in Economics]. https://ideas.repec.org/p/anp/en2013/128.html
Cajueiro, D. O., Divino, J. A. & Tabak, B. M. (2009). Forecasting the yield curve for Brazil, Working Papers Series 197, Banco Central do Brasil. https://ideas.repec.org/p/bcb/wpaper/197.html
Caldeira, J., Moura, G. V. & Portugal, M. S. (2010). Efficient yield curve estimation and forecasting in Brazil, Revista EconomiA, 11(1), 27–51. https://www.anpec.org.br/revista/vol11/vol11n1p27_51.pdf
Campbell, J. Y. (1986). Bond and stock returns in a simple exchange model, The Quarterly Journal of Economics, 101(4), 785–803. https://doi.org/10.2307/1884178
Cox, J. C., Ingersoll, J. E. & Ross, S. A. (1985). A theory of the term structure of interest rates, Econometrica, 53(2), 385–407. https://doi.org/10.2307/1911242
Diebold, F. X. & Li, C. (2006). Forecasting the term structure of government bond yields, Journal of Econometrics, 130(2), 337–364. https://doi.org/10.1016/j.jeconom.2005.03.005
Elman, J. L. (1990). Finding structure in time, Cognitive Science, 14(2), 179–211. https://doi.org/10.1207/s15516709cog1402_1
Faria, L. G. (2024). Aprendendo a curva de juros brasileira, Brazilian Review of Finance, 22(1), 1–29. https://doi.org/10.12660/rbfin.v22n1.2024.89588
Fernandes, M., Nunes, C. & Reis, Y. (2021). What drives the nominal yield curve in Brazil?, Brazilian Review of Econometrics, 40(2), 267–284. https://doi.org/10.12660/bre.v40n22020.79438
Filipović, D., Pelger, M. & Ye, Y. (2025). Stripping the discount curve - A robust machine learning approach, Management Science, 0(0). https://doi.org/10.1287/mnsc.2023.01401
Fischer, T. & Krauss, C. (2018). Deep learning with long short-term memory networks for financial market predictions, European Journal of Operational Research, 270(2), 654–669. https://doi.org/10.1016/j.ejor.2017.11.054
Garza, A., Mergenthaler Canseco, M., Challú, C. & Olivares, K. G. (2022). StatsForecast: Lightning fast forecasting with statistical and econometric models, PyCon Salt Lake City, Utah, US 2022.
Goodfellow, I., Bengio, Y. & Courville, A. (2016). Deep Learning, MIT Press. https://www.deeplearningbook.org
Hochreiter, S. & Schmidhuber, J. (1997). Long short-term memory, Neural Computation, 9(8), 1735–1780. https://doi.org/10.1162/neco.1997.9.8.1735
Hyndman, R. J., Athanasopoulos, G., Garza, A., Challu, C., Mergenthaler, M. & Olivares, K. G. (2025). Forecasting: Principles and Practice, the Pythonic Way, OTexts, Melbourne, Australia. Online edition. Accessed on 3 January 2026. https://otexts.com/fpppy/
Nelson, C. R. & Siegel, A. F. (1987). Parsimonious modeling of yield curves, The Journal of Business, 60(4), 473–89. https://doi.org/10.1086/296409
Olivares, K. G., Challú, C., Garza, A., Canseco, M. M. & Dubrawski, A. (2022). NeuralForecast: User friendly state-of-the-art neural forecasting models, PyCon Salt Lake City, Utah, US 2022. https://github.com/Nixtla/neuralforecast
Paiva, E. & Pereira, F. (2022). Deep Learning para Processamento de Linguagem Natural, SBC, pp. 21–50. https://doi.org/10.5753/sbc.8690.0.2
Rosa, P. S. & Gartner, I. R. (2018). Financial distress em bancos brasileiros: Um modelo de alerta antecipado, Revista Contabilidade & Finanças, 29(77), 312–331. https://doi.org/10.1590/1808-057x201803910
Sima, D. (2022). Reconstruction-based Anomaly Detection on Financial Market Data, PhD thesis, Politecnico Milano. Accessed: 17 Sep. 2025. https://hdl.handle.net/10589/215104
Suimon, Y., Sakaji, H., Izumi, K. & Matsushima, H. (2020). Autoencoder-based three-factor model for the yield curve of Japanese government bonds and a trading strategy, Journal of Risk and Financial Management, 13(4), 82. https://doi.org/10.3390/jrfm13040082
Svensson, L. E. O. (1994). Estimating and interpreting forward interest rates: Sweden 1992 - 1994, Working Paper 4871, National Bureau of Economic Research. https://doi.org/10.3386/w4871
Urolagin, S., Sharma, N. & Datta, T. K. (2021). A combined architecture of multivariate LSTM with Mahalanobis and Z-score transformations for oil price forecasting, Energy, 231, 120963. https://doi.org/10.1016/j.energy.2021.120963
Vasicek, O. (1977). An equilibrium characterization of the term structure, Journal of Financial Economics, 5(2), 177–188. https://doi.org/10.1016/0304-405X(77)90016-2