"Forecasting Interest Rates using Vasicek's Term Structure Model"(in Japanese)
Takao Kobayashi, Akihiro Ochi
Abstract
Takao Kobayashi, Akihiro Ochi
Abstract
We compare two approaches to forecast future interest rates; (1) naive approach based on "the unbiased expectations hypothesis", and (2) an approach that incorporates investors' risk aversion. In particular, we use Vasicek's one-factor term structure model, which is the simplest model of interest rate movements that incorporates risk aversion of market participants and mean reversion of interest rates. The main conclusion is two-fold. Firstly, the empirical evidence shows that there is a strong mean reversion in the Japanese interest rates and that its speed is much more sluggish than the US interest rates. Secondly, the use of Vasicek's model did not improve the predictive power of current market data significantly. The evidence strongly suggests the need to use a multiple-factor model to improve the fit to the observed yield curve and the predictive ability of the model for future interest rate movements, if the market data contains any information.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
We compare two approaches to forecast future interest rates; (1) naive approach based on "the unbiased expectations hypothesis", and (2) an approach that incorporates investors' risk aversion. In particular, we use Vasicek's one-factor term structure model, which is the simplest model of interest rate movements that incorporates risk aversion of market participants and mean reversion of interest rates. The main conclusion is two-fold. Firstly, the empirical evidence shows that there is a strong mean reversion in the Japanese interest rates and that its speed is much more sluggish than the US interest rates. Secondly, the use of Vasicek's model did not improve the predictive power of current market data significantly. The evidence strongly suggests the need to use a multiple-factor model to improve the fit to the observed yield curve and the predictive ability of the model for future interest rate movements, if the market data contains any information.
Key concepts: Vasicek model, Yield curve, Interest rate, Econometrics, Mean reversion, Economics, Term (time), Affine term structure model