We develop a new likelihood-based approach to signing trades in the absence of quotes. This approach is equally efficient as the existing Markov-chain Monte Carlo methods, but more than ten times faster. It can address the occurrence of multiple trades at the same time and allows for analysis of...