Gah!28 April 2009
Another one of the many perturbing errors in Jeffrey's Subjective Probability, this from §4.1:
If I am sure that it was one of the years from 1951 to 1959, with equal probabilities, then my ex(X) will beOne doesn't even have to do any arithmetic (I didn't) to spot the error; for a sequence of an odd number of equally-spaced integers, the average is the middle-most value, in this case 1955.
9 + … +
9, which works out to be 1954.5.