Calibrating Rate Models After Negative Rates Emerge

Summary

The document raises a calibration question for models of negative interest rates. It suggests that the magnitude of negative rates may move with fixed-asset prices and cross-currency basis spreads, then asks whether volatility and correlation estimates fitted to earlier, nonnegative rate observations remain dependable once rates turn negative.

One proposed alternative is to fit the model using a shorter sample beginning when negative rates were observed in the market, while distinguishing observed negative rates from negative forward rates. The text offers no empirical comparison, model specification, or answer about which sample is preferable. Its central issue is the trade-off between relying on a longer history that may describe a different regime and using a limited number of observations that may provide weak estimates. Any calibration choice would need to account for the instrument, market, and period under study.

Key ideas

• The document questions whether pre-negative-rate volatility estimates transfer to a negative-rate regime.
• It proposes examining relationships between negative rates, fixed-asset prices, and cross-currency basis spreads.
• A shorter sample focused on observed negative rates may be more relevant but contains less data.
• Observed negative rates and negative forward rates are treated as distinct calibration considerations.
• The document poses these issues without providing data or a preferred estimation method.

Tags

# Neglect the positive values in negative interest rates modelling?


# Neglect the positive values in negative interest rates modelling?












The magnitude of the negative interested rate should vary correlated with the increase in fixed assets prices and with cross-currency basis spreads.

Could their volatility / correlation coefficients, as historically fitted for all the previous non-negative observed interest rates be reliable and considered in backtesting the model?

Would it not be better to consider low data sampling analysis and fit them only since negative interest rates (not negative forward ones) have been observed on the market?

Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.