Hidden Markov models for binary time series
For the present still working with the assumption of Markan priority, we now consider an alternative approach to the one adopted in Chapter 3 for modelling the binary time series (xt) and (yt) that represent Matthew’s and Luke’s verbal agreements with Mark, respectively. The underlying idea is that, when Mark was being used as a source by either Matthew or Luke, different parts of Mark’s text were handled in different ways. For example, some sections of Mark’s text were followed quite closely, almost word for word, but other sections were followed rather loosely. In some sections of text, Mark might have been essentially the only source for Matthew or Luke, whereas in other sections substantial use of additional sources, including oral tradition, might have been made. Furthermore, some sections of Mark’s text appear not to have been used at all. For example, the Markan pericope Mk 8:22-26, “The Blind Man of Bethsaida”, is omitted by both Matthew and Luke, and, most strikingly of all, there is Luke’s great omission, where he has omitted all the Markan pericopes in the section of text from Mk 6:45-8:10.