The information in the term structure: A non-parametric investigation
β Scribed by Bruce Mizrach
- Publisher
- John Wiley and Sons
- Year
- 1996
- Tongue
- English
- Weight
- 993 KB
- Volume
- 15
- Category
- Article
- ISSN
- 0277-6693
No coin nor oath required. For personal study only.
β¦ Synopsis
This paper proposes a non-linear term structure model that nests the discrete and continuous time models as special cases. I estimate the model non-parametrically using nearest-neighbours regression. In sample, the non-linear model matches the standard theories, but out of sample, it offers substantial improvement. Linear models fail to track future interest rates: a random walk dominates the forward rate as a predictor for 3-month Treasury bills. A non-linear forecast based on the spread is shown statistically to be the best forecast.
π SIMILAR VOLUMES
The authors would like to thank James Weston, Michael Highfield, and participants at 2003 meeting of the Southern Finance Association. We would also like to thank an anonymous referee whose comments substantially improved this article. The College of Business, James Madison University, supported thi