One of the hardest tasks of a public key infrastructure (PKI) is to manage revocation. Newcommunication paradigms push the revocation system to the limit and an accurate resource assessmentis necessary before implementing a particular revocation distribution system. In this context, a precisemodeling of certificate revocation is necessary. In this article, we analyze empirical data from realCAs to develop an accurate and rigorous model for certificate revocation. One of the key findings ofour analysis is that the certificate revocation process is statistically self-similar. The proposed modelis based on an autoregressive fractionally integrated moving average (ARFIMA) process. Then, usingthis model, we show how to build a synthetic revocation generator that can be used in simulationsfor resource assessment. Finally, we also show that our model produces synthetic revocation tracesthat are indistinguishable for practical purposes from those corresponding to actual revocations.