In this paper we compare the out of sample forecasts from four alternative interest rate models based 00 expanding information sets. The random walk model is the most restrictive. The univariate time series model allows for a richer dynamic pattern and more conditioning information on own rates. The multivariate time series model permits a flexible dynamic pattern with own-and cross-series information. Finally, the forecasts from the MPS econometric model depend on the full model structure and information set.
In theory, more information is preferred to less. In practice, complicated misspecified models can perform much worse than simple (also probably misspecified) models. For forecasts evaluated over the volatile 1970s the multivariate time series model forecasts are considerably better than those from simpler models which use less conditioning information, as well as forecasts from the MPS model which uses substantially more conditioning information but also imposes 'structural' economic restrictions. KEY WOKDS Interest rates Time series forecasts Model comparisons Recent theoretical work (Begg (1984) and Grossman and Stiglitz (1976, 1980 for example) suggests that the random walk model of interest rates may be an over simplification for a number of reasons, including conditioning information and information costs. We examine the out of sample short-term and long-term interest rate forecasting performance of the random walk model and three alternatives that use successively more information: (i) univariate time series models, (ii) a multivariate time series model, and (iii) a large econometric model (the MIT-Penn-SSRC (MPS) model). The MPS model was chosen because it is a large structural model with a sophisticated financial sector used by the Federal Reserve Board for policy purposes, on which a significant amount of resources is expended.
Monthly forecasts of a short rate (the 4 to 6 month commercial paper rate) and a long rate (Moody's BAA corporate bond rate) are evaluated out of sample over the period 1971-8 to 1977-10. This was a period of major interest rate movements that include the wage-price freeze, OPEC embargo and subsequent price rise, and a severe economic downturn; as such, it provides a critical test of the ability of models to forecast when good forecasts are most needed,