A General Derivative Identity for the Conditional Expectation with Focus on the Exponential Family

arXiv2105.05106MaRDI QIDQ6367386FDOQ6367386

Publication date: 11 May 2021

Abstract: Consider a pair of random vectors

(m a t h b f X, m a t h b f Y)

and the conditional expectation operator

m a t h b b E [m a t h b f X | m a t h b f Y = m a t h b f y]

. This work studies analytic properties of the conditional expectation by characterizing various derivative identities. The paper consists of two parts. In the first part of the paper, a general derivative identity for the conditional expectation is derived. Specifically, for the Markov chain

m a t h b f U l e f t r i g h t a r r o w m a t h b f X l e f t r i g h t a r r o w m a t h b f Y

, a compact expression for the Jacobian matrix of

m a t h b b E [m a t h b f U | m a t h b f Y = m a t h b f y]

is derived. In the second part of the paper, the main identity is specialized to the exponential family. Moreover, via various choices of the random vector

m a t h b f U

, the new identity is used to recover and generalize several known identities and derive some new ones. As a first example, a connection between the Jacobian of

m a t h b b E [m a t h b f X | m a t h b f Y = m a t h b f y]

and the conditional variance is established. As a second example, a recursive expression between higher order conditional expectations is found, which is shown to lead to a generalization of the Tweedy's identity. Finally, as a third example, it is shown that the

k

-th order derivative of the conditional expectation is proportional to the

(k + 1)

-th order conditional cumulant.

This page was built for publication: A General Derivative Identity for the Conditional Expectation with Focus on the Exponential Family

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6367386)