8Extensions of the method of maximum entropy
In this chapter we consider two extensions of the maximum entropy method to cover two somewhat related cases:
(1)There may be errors in the data. This means that the moments μi that appear in the right-hand side of (7.1)–(7.2) are exactly known, but up to errors in the measurement.
(2)The moments μi are not known. What is known is an interval, say [ai, bi], in which the moment μi falls.
(3)There is error in the measurement and the data is specified to lie in an interval.
Let us state the three possible generalized moment problems. The first one consists of finding a probability density f of an unknown probability P with respect to a reference measure m on a sample space (Ω, ℱ), and to estimate a ...
Get Loss Data Analysis now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
