Olga Avdeeva1, Alexander Tsyplakov1
  • 1 Novosibirsk State University, 2, Pirogova st., Novosibirsk, 630090, Russian Federation

A Method for Adaptive Estimation of the Term Structure of Interest Rates

2015. Vol. 19. No. 4. P. 609–639 [issue contents]

One source of information for analyzing profitability of investments is the yields on go­vernment bonds with different maturities. However, the fact that bonds are rarely zero-coupon prevents the direct use of these yields. Since payments on bonds are done indifferent periods, the price of a bond is a sum of nonlinear components with interest rates corresponding to dif­ferent terms to maturity. The problem of obtaining zero-coupon yield data from bond prices is most acute for the markets such as Russian, where trade volumes are not as large and trans­actions do not happen very often.

The paper proposes a relatively easy-to-implement method for constructing parametric zero-coupon yield curve given price data, in which the parameters of the curve and volatility of the random disturbances are assumed to be time-varying, while the distribution of random disturbances is fat-tailed. The method modifies the classical Kalman filter and is based on the score vector of the measurement equation and the corresponding information matrix. The closest analogues of the proposed method are GAS (generalized autoregressive score) and DSC (dynamic conditional score) approaches . The method is not demanding to the quality of the data, since it is robust to outliers. Estimates of the interest rate curve are obtained adaptively and reflect the current market situation. Performance of the method is assessed against the data on the Russian government bond market for the period from January 2008 to April 2015. We obtain estimates of the parameters of the dynamic Nelson – Siegel curve for this period.
Citation: Avdeeva O., Tsyplakov A. (2015) Metod adaptivnogo otsenivaniya srochnoy struktury protsentnykh stavok [A Method for Adaptive Estimation of the Term Structure of Interest Rates]. HSE Economic Journal, vol. 19, no 4, pp. 609-639 (in Russian)
Rambler's Top100 rss