Paper: arXiv 2506.06317

Authors: Amit Kumar Jha

Abstract

This study is motivated by empirical observations of periodic fluctuations in interest rates, notably long-term economic cycles spanning decades, which the conventional Hull-White short-rate model fails to adequately capture. To address this limitation, we propose an extension that incorporates a sinusoidal, time-varying mean reversion speed, allowing the model to reflect cyclic interest rate dynamics more effectively. The model is calibrated using a comprehensive dataset of daily U.S. Treasury yield curves obtained from the Federal Reserve Economic Data (FRED) database, covering the period from January 1990 to December 2022. The dataset includes tenors of 1, 2, 3, 5, 7, 10, 20, and 30 years, with the most recent yields ranging from 1.22% (1-year) to 2.36% (30-year). Calibration is performed using the Nelder-Mead optimization algorithm, and Monte Carlo simulations with 200 paths and a time step of 0.05 years. The resulting 30-year zero-coupon bond price under the proposed model is 0.43, compared to 0.47 under the standard Hull-White model. This corresponds to root mean squared errors of 0.12% and 0.14%, respectively, indicating a noticeable improvement in fit, particularly for longer maturities. These results highlight the model’s enhanced capability to capture long-term yield dynamics and suggest significant implications for bond pricing, interest rate risk management, and the valuation of interest rate derivatives. The findings also open avenues for further research into stochastic periodicity and alternative interest rate modeling frameworks.

Complexity vs Empirical Score

  • Math Complexity: 7.0/10
  • Empirical Rigor: 6.5/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper introduces a novel extension to the Hull-White model with a sinusoidal time-varying parameter, involving advanced SDEs, Fourier analysis, and Monte Carlo simulations, indicating high mathematical density. It is supported by comprehensive real-world data from FRED, calibration using Nelder-Mead optimization, and backtest-ready metrics like RMSE, demonstrating strong empirical implementation.

Research Flowchart

  flowchart TD
  A["Research Goal: <br>Model Long-Term Periodicity in Interest Rates"] --> B{"Methodology: <br>Extended Hull-White Model"}
  B --> C["Data Inputs: <br>U.S. Treasury Yields (1990-2022)"]
  C --> D["Calibration: <br>Nelder-Mead Optimization"]
  D --> E["Computational Process: <br>Monte Carlo Simulation"]
  E --> F{"Key Outcomes"}
  F --> G["RMSE: 0.12% <br>(Improved Fit vs. 0.14%)"]
  F --> H["Bond Price: $0.43 <br>(30-Year Zero-Coupon)"]
  F --> I["Implications: <br>Bond Pricing & Risk Management"]

Evidence Card

What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card

Proposes

The paper proposes a sinusoidal extension to the Hull‑White model that adds a time‑varying mean reversion speed to capture long‑term cyclic interest‑rate dynamics.

Source sentence
we propose an extension that incorporates a sinusoidal, time-varying mean reversion speed, allowing the model to reflect cyclic interest rate dynamics more effectively.

Assumes 4

  • Mean reversion speed follows κt = κ0 + A sin(ωt). model
    Source sentence
    κt = κ0 + A sin(ωt), where κ0 is the baseline speed, A is the amplitude, and ω governs the frequency of oscillation.
  • Short‑rate dynamics follow the standard Hull‑White SDE under the risk‑neutral measure. model
    Source sentence
    drt = κ(θ −rt)dt + σdWt, where rt is the short rate at time t, κ is the constant mean reversion speed, θ is the long‑term mean rate, σ is the volatility, and dWt is a standard Wiener process under the risk‑neutral measure.
  • Yield data are non‑stationary but exhibit significant autocorrelation. statistical
    Source sentence
    The Augmented Dickey‑Fuller (ADF) test results, presented in Table 2, indicate p‑values ranging from 0.568 (30‑year) to 0.897 (1‑year), all exceeding the 0.05 threshold, suggesting that the yield series are non‑stationary.
  • Monte Carlo simulation with 200 paths and a time step of 0.05 years provides accurate pricing for the sinusoidal extension. other
    Source sentence
    Monte Carlo simulations with 200 paths and a time step of 0.05 years.

Data

  • Daily U.S. Treasury yield curves · Federal Reserve Economic Data (FRED) macro public
    January 1990 to December 2022 · Tenors of 1, 2, 3, 5, 7, 10, 20, and 30 years
    Source sentence
    The model is calibrated using a comprehensive dataset of daily U.S. Treasury yield curves obtained from the Federal Reserve Economic Data (FRED) database, covering the period from January 1990 to December 2022. The dataset includes tenors of 1, 2, 3, 5, 7, 10, 20, and 30 years.

Recognised sources: FRED

Costs

Not applicable The paper reports no trading or execution result.

Finds 3

  • The sinusoidal model yields a 30‑year zero‑coupon bond price of 0.43, closer to the observed price of 0.4926 than the standard model’s 0.47, with RMSE 0.12% versus 0.14% for the standard model. pricing-calibration
    Source sentence
    The resulting 30‑year zero‑coupon bond price under the proposed model is 0.43, compared to 0.47 under the standard Hull‑White model. This corresponds to root mean squared errors of 0.12% and 0.14%, respectively.
  • Fourier Transform analysis identifies dominant periodicities of approximately 3.7, 5.5, 11, and 22 years, with the 22‑year cycle most consistent across tenors. statistical
    Source sentence
    A Fourier Transform (FT) analysis of the yield data reveals dominant periodicities of approximately 3.7, 5.5, 11, and 22 years, with the 22‑year cycle being the most consistent across all tenors.
  • Calibration using the Nelder‑Mead algorithm and Monte Carlo simulation provides robust parameter estimates across a 33‑year historical dataset. theoretical
    Source sentence
    Calibration is performed using the Nelder‑Mead optimization algorithm, and Monte Carlo simulations with 200 paths and a time step of 0.05 years.

Checked by others

No other paper in the archive has been found to support or contradict these findings (compared with 21 related papers). Coverage is partial, so this is neither replication nor refutation.

Code

No repository by the authors found in the paper.

Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.

Cited References (19)

  1. Pricing Interest-Rate-Derivative Securities (1990) — link
  2. Interest Rate Models - Theory and Practice: With Smile, Inflation and Credit (2006)
  3. Options, Futures, and Other Derivatives (2006)
  4. The Econometrics of Financial Markets (1997)
  5. Interest Rates under Falling Stars (2018) — link
  6. U.S. Treasury Yields (2022)
  7. Numerical Procedures for Implementing Term Structure Models II: Two-Factor Models (1994) — link
  8. Bond Pricing and the Term Structure of Interest Rates: A New Methodology for Contingent Claims Valuation (1992) — link
and 11 more
  1. Specification Analysis of Affine Term Structure Models (2000) — link
  2. Which Moments to Match? (1996) — link
  3. Recent Dynamics of G7 Bond Yields: A Spectral Perspective (2006) — link
  4. A Theory of the Term Structure of Interest Rates (1985) — link
  5. A New Approach to the Economic Analysis of Nonstationary Time Series and the Business Cycle (1989) — link
  6. A Yield-Factor Model of Interest Rates (2000) — link
  7. Regime Switches in Interest Rate Volatility (2008) — link
  8. Forecasting the Term Structure of Government Bond Yields (2006) — link
  9. Interest Rate Models - Theory and Practice: With Smile, Inflation and Credit (2007)
  10. Bayesian Forecasting and Dynamic Models (1997)
  11. Common Factors Affecting Bond Returns (1991) — link

0 of 19 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.