Take in the generator domain . The semigroup commutation with its unbounded generator implies and
Indeed, applying the bounded operator to the difference quotient at zero gives the right derivative ; strong continuity gives the full orbit derivative for . Now integration by parts in the Bochner integral yields
The boundary term at infinity vanishes because . The derivative integral exists as a Bochner integral, since its integrand is bounded in norm by .
Multiplying this identity by gives
Part (b) applies to the fixed vector , so
The bounded operators are the Yosida approximation of the semigroup generator. The proof only applies the unbounded operator to ; the contraction and limit arguments then act on the ordinary vector .
The convergence is in the strong operator topology: for each fixed , one must prove . First, the contraction semigroup bound gives
Thus is a contraction. This is the useful scaling of the contraction estimate in the source's hint. Since ,
Fix . By strong continuity, choose so that for . Split the integral there and use for the tail:
Let and then . Therefore
The distinction from operator norm convergence is essential. For example, the multiplication semigroup on defined by is a strongly continuous semigroup of contractions: coordinatewise convergence and dominated convergence applied to prove strong continuity. Its resolvent operator is , and
Thus the claimed convergence cannot in general be read as convergence in operator norm.
The diffusion generator is . By the Feynman-Kac formula, the required conditional characteristic function solves
Use separation of variables in the form . Since and , the temporal factor satisfies the ordinary differential equation
Thus , and
This is bounded in absolute value by one, so it lies in the uniqueness class of the Feynman-Kac formula. As an independent check, integrating the stochastic differential equation gives . The independent Brownian increments have the normal distribution with mean zero and variance , yielding exactly the same characteristic function. For , the conditional distribution is a point mass at , and the same boxed expression still applies.
For the stochastic differential equation , the diffusion generator on smooth functions is
One form of the Feynman-Kac formula is the following. Suppose the stochastic differential equation is well posed and nonexplosive, is bounded, the terminal data and source are bounded, and is a classical solution bounded on finite time intervals of
With the regularity needed for the Itô formula, the parabolic Feynman-Kac formula with a source states
Here means expectation for the diffusion process started at . Under hypotheses ensuring existence of a classical solution, this representation also identifies that solution uniquely in the stated bounded class. In the remaining parts, , so solves the terminal-value Kolmogorov backward equation . Complex terminal data are allowed by applying the real formula to their real part and imaginary part separately.
The displayed identity requires a correction to the process construction in the source. In a piecewise-deterministic Markov process, the instantaneous hazard function is evaluated at the current flowing state , and the destination is sampled from . Keeping the jump times and destinations of the original Markov jump process instead evaluates both quantities at its last reset state , which generally differs from . The source PDF confirms that these original jump times and destinations were retained. Consequently its definition does not in general produce the asserted transition kernel, or even a Markov process in the state variable alone.
For an explicit counterexample, take , , , and let every be the standard normal distribution, independently of . These are smooth coefficients with bounded positive jump rate, so nonexplosion holds. In the literal construction, the first holding time has the exponential distribution of rate . Thus the probability distribution of has an atom of mass at . Paths with at least one jump contribute no atom there, because the reset has a normal density. The asserted identity instead assigns to this no-jump atom the mass
which differs for every .
Here is the full derivation for the intended piecewise-deterministic Markov process. Assume nonexplosion; smoothness alone does not guarantee it. Start at , let be the first jump after , and set
One precise construction draws an independent unit-rate exponential random variable and takes , with . This implements a hazard function along the flow of a vector field, rather than a fixed exponential holding time at the last reset. Its survival probability and jump-time probability density function are
Equivalently, with . On , the terminal state is . Conditional on , draw the new state from and restart the same construction from . The Markov property and the tower property of conditional expectation therefore give the first-jump decomposition of a piecewise-deterministic Markov process
This proves the requested identity under the corrected construction, for every bounded Borel measurable function . The two contributions partition according to whether the first jump occurs before ; later jumps are already contained in .
For comparison, the literal construction does have a different renewal identity when started at a reset with . If denotes only these reset-start operators, the same conditional expectation argument gives
These need not be the transition operators conditional on at an arbitrary time: that observation does not specify the hidden reset state. This distinction explains exactly why replacing only the between-jump paths is insufficient.
Let be a Markov chain with transition kernel , and let be an independent Poisson process of constant rate . A Pseudo-Poisson process is . On the Banach space of bounded Borel measurable functions, with the supremum norm, write
Its infinitesimal generator, transition semigroup, and resolvent operator are respectively
and, for ,
The transition semigroup averages against the Poisson distribution of the number of updates. Since is a Markov operator, ; both series converge in operator norm. Updates with are allowed and do not count as changes of the observed state. The degenerate rate gives , , and .
Write and . Conditional independence of the next count given allows us to average its Poisson distribution over the posterior distribution:
This Poisson-gamma mixture is the posterior predictive distribution. In the stated negative binomial distribution convention its parameters are
Its expected value equals the posterior mean of , while its variance includes both conditional Poisson variation and uncertainty in .
Let . The continuous uniform distribution likelihood contributes , while the Pareto distribution prior contributes . Their product proves Uniform-Pareto conjugacy:
Thus the posterior distribution is Pareto with shape and lower bound . Its density is strictly decreasing on its support, so the highest posterior density interval begins at . Since the upper-tail probability is , the required upper endpoint satisfies . The Pareto highest density interval is therefore

There are unlisted articles, also show them or only show them.