Empirical tests of the Longstaff extendible warrant model

In the 1970s and 1980s more than 20% of the firms extended their warrants' expiration date. This study uses a sample of over 22500 warrant price observations to test the pricing performance of the Longstaff writer-extendible warrant model. The Longstaff model is found to generate lower absolute pricing errors than the benchmark Black-Scholes model, and the pattern of the pricing error reduction fits the implications of Longstaff (1990). The pricing performance of the Longstaff model is particularly impressive for warrants which were eventually extended. This suggests that investors are aware and take into account the possibility of extension.