DURATION MOMENTS AND YIELD CURVE MOVEMENTS

In this paper we will analyze the relationship between the value and duration moments of a cash flow and movements in the yield curve. We will show that for changes in the yield curve which can be related to tⁿ , the 1st order changes in the net present value of a cash flow are linearly dependent on the n + lth duration moments, and that the 2nd order changes are dependent on the sum of duration moments of order 2 n + 1 and 2 n + 2. We will use this relationship to tilt tracking portfolios so as to protect them against specific changes in the yield curve.     

Significant Issues for the Future of Product Innovation

The value of a technology strategy has been increasingly discussed by R&D personnel and those involved in product development and business planning efforts. In this issue Albert Rubenstein and Americo Albala elevate this discussion by showing the importance of preparing a firm to compete by stressing the development of important, relevant technologies. Discovery and development must be channeled into areas that will be needed to support business initiatives likely to occur in the future. Albala stresses the importance of this change for the development of economies in nations that have been disappointed by the lack of success of importing technologies from more developed nations, while Rubenstein maintains that the pace of marketplace change requires a technology strategy in order to reduce the misapplication of scarce resources within the firm and the possibility of being blindsided by competitive developments. These essays continue a year-long series of contributions that the editor-in-chief solicited from members of the editorial board. Members were asked to reflect upon changes and opportunities that they see influencing our profession during the coming decade. Both of these short essays are designed to introduce new perspectives. It is not essential that you agree with the recommendations, but we hope that you are stimulated as you reflect on the issues they raise.     

MODELING THE TERM STRUCTURE FROM THE ON‐THE‐RUN TREASURY YIELD CURVE

We propose a new model to estimate the term structure of interest rates using observed on‐the‐run Treasury yields. The new model is an improvement over models that require a priori knowledge of the shape of the yield curve to estimate the term structure. The general form of the model is an exponential function that depends on the estimation of four parameters fit by nonlinear least squares and has straightforward interpretations. In comparing the proposed model with current yield‐curve‐smoothing models, we find that, for the data used, the proposed model does best overall in terms of pricing accuracy both in sample and out of sample. JEL classification: E43, G12     

EFFECT OF FED POLICY ACTIONS ON THE DEFAULT LIKELIHOOD OF COMMERCIAL BANKS

Using a measure of default likelihood based on an option pricing method, we provide evidence that Fed policy actions affect the financial distress of commercial banks. When the Fed increases (decreases) interest rates, the measure of default likelihood increases (decreases). We show that when the Fed uses a tight money policy, the increase in default likelihood is more pronounced for banks that have less capital, have greater financial leverage, are smaller, have fewer growth opportunities, and have lower asset quality. Additionally, the effects on bank default likelihood are more pronounced when the Fed's policy signals less concern about economic growth, as indicated by its bias toward further tightening, and when there is a market expectation of higher short‐term market rates in the future.     

VAR,ERROR CORRECTION AND PRETEST FORECASTS AT LONG HORIZONS

This paper focuses on the construction of forecasts over long horizons where a typical long-horizon forecast might span four years using 20 to 40 years’ data. It is argued that the presence of persistence in the form of unit or near-unit autoregressive roots poses substantial difficulties for long-horizon interval and point forecasting. These difficulties may not be overcome even by efficient pre-testing or model-selection procedures and might, in general, lead to point forecasts with large asymptotic root mean squared errors and undesirably wide prediction intervals.