[Mutual fund attrition can create problems for a researcher because funds that disappear tend to do so due to poor performance. In this article we estimate the size of the bias by tracking all funds that existed at the end of 1976. When a fund merges we calculate the return, taking into account the merger terms. This allows a precise estimate of survivorship bias. In addition, we examine characteristics of both mutual funds that merge and their partner funds. Estimates of survivorship bias over different horizons and using different models to evaluate performance are provided.]

Mutual fund attrition can create problems for a researcher because funds that disappear tend to do so due to poor performance. In this article we estimate the size of the bias by tracking all funds that existed at the end of 1976. When a fund merges we calculate the return, taking into account the merger terms. This allows a precise estimate of survivorship bias. In addition, we examine characteristics of both mutual funds that merge and their partner funds. Estimates of survivorship bias over different horizons and using different models to evaluate performance are provided.

This paper examines asset pricing theories for treasury bonds using longer maturities than previous studies and employing a simple multi-factor model. We allow bond factor loadings to vary over time according to term structure variables. The model examines not only the time variation in the expected returns of bonds but also their unexpected returns. This allows us to explicitly test some asset pricing restrictions which are difficult to study under existing frameworks. We confirm that the pure expectation theory of the term structure of interest rates is rejected by the data. Our empirical study of a two-factor model finds substantial evidence of time-varying term-premiums and factor loadings. The fact that factor loadings vary with long-term interest rates and yield spreads suggest that bond return volatilities are sensitive to interest rate levels.