This article is about infinitesimal generator for general stochastic processes. For generators for the special case of finite-state continuous time Markov chains, see
transition rate matrix.
In mathematics — specifically, in stochastic analysis — the infinitesimal generator of a Feller process (i.e. a continuous-time Markov process satisfying certain regularity conditions) is a Fourier multiplier operator[1] that encodes a great deal of information about the process.
The generator is used in evolution equations such as the Kolmogorov backward equation, which describes the evolution of statistics of the process; its L2 Hermitian adjoint is used in evolution equations such as the Fokker–Planck equation, also known as Kolmogorov forward equation, which describes the evolution of the probability density functions of the process.
The Kolmogorov forward equation in the notation is just
, where
is the probability density function, and
is the adjoint of the infinitesimal generator of the underlying stochastic process. The Klein–Kramers equation is a special case of that.
Definition[edit]
General case[edit]
For a Feller process
with Feller semigroup
and state space
we define the generator[1]
by
![{\displaystyle D(A)=\left\{f\in C_{0}(E):\lim _{t\downarrow 0}{\frac {T_{t}f-f}{t}}{\text{ exists as uniform limit}}\right\},}](https://wikimedia.org/api/rest_v1/media/math/render/svg/0b24ec54d1e2f117bd58d4af6c84efe1a7ad4bb1)
![{\displaystyle Af=\lim _{t\downarrow 0}{\frac {T_{t}f-f}{t}},~~{\text{ for any }}f\in D(A).}](https://wikimedia.org/api/rest_v1/media/math/render/svg/25c089416d3c51eb27f72c5b868881d33d397c6a)
Here
![{\displaystyle C_{0}(E)}](https://wikimedia.org/api/rest_v1/media/math/render/svg/084c75ea3c3c8ca5d17c85c8deda3e34cf39e26d)
denotes the Banach space of continuous functions on
![{\displaystyle E}](https://wikimedia.org/api/rest_v1/media/math/render/svg/4232c9de2ee3eec0a9c0a19b15ab92daa6223f9b)
vanishing at infinity, equipped with the supremum norm, and
![{\displaystyle T_{t}f(x)=\mathbb {E} ^{x}f(X_{t})=\mathbb {E} (f(X_{t})|X_{0}=x)}](https://wikimedia.org/api/rest_v1/media/math/render/svg/d249e365a8b3b2d4444d91197954e5597c1a606f)
. In general, it is not easy to describe the domain of the Feller generator. However, the Feller generator is always closed and densely defined. If
![{\displaystyle X}](https://wikimedia.org/api/rest_v1/media/math/render/svg/68baa052181f707c662844a465bfeeb135e82bab)
is
![{\displaystyle \mathbb {R} ^{d}}](https://wikimedia.org/api/rest_v1/media/math/render/svg/a713426956296f1668fce772df3c60b9dde8a685)
-valued and
![{\displaystyle D(A)}](https://wikimedia.org/api/rest_v1/media/math/render/svg/47f833d059e4565ca5c84985c780b21f1f89f0b9)
contains the
test functions (compactly supported smooth functions) then
[1]
![{\displaystyle Af(x)=-c(x)f(x)+l(x)\cdot \nabla f(x)+{\frac {1}{2}}{\text{div}}Q(x)\nabla f(x)+\int _{\mathbb {R} ^{d}\setminus {\{0\}}}\left(f(x+y)-f(x)-\nabla f(x)\cdot y\chi (|y|)\right)N(x,dy),}](https://wikimedia.org/api/rest_v1/media/math/render/svg/09d32fadc20e8f4004693bc638259dfa5467c8f8)
where
![{\displaystyle c(x)\geq 0}](https://wikimedia.org/api/rest_v1/media/math/render/svg/0795969f536b34602a9e4ac824811e7691577b86)
, and
![{\displaystyle (l(x),Q(x),N(x,\cdot ))}](https://wikimedia.org/api/rest_v1/media/math/render/svg/fcbf952fcefbc1f71f502f74faa9d8fc53e198b2)
is a
Lévy triplet for fixed
![{\displaystyle x\in \mathbb {R} ^{d}}](https://wikimedia.org/api/rest_v1/media/math/render/svg/f351538c1465ec3881164b501f612b1f54cbfe7e)
.
Lévy processes[edit]
The generator of Lévy semigroup is of the form
![{\displaystyle Af(x)=l\cdot \nabla f(x)+{\frac {1}{2}}{\text{div}}Q\nabla f(x)+\int _{\mathbb {R} ^{d}\setminus {\{0\}}}\left(f(x+y)-f(x)-\nabla f(x)\cdot y\chi (|y|)\right)\nu (dy)}](https://wikimedia.org/api/rest_v1/media/math/render/svg/12d9b181c6961d8e9e58325e3307805d806ce5b8)
where
![{\displaystyle l\in \mathbb {R} ^{d},Q\in \mathbb {R} ^{d\times d}}](https://wikimedia.org/api/rest_v1/media/math/render/svg/d44e8cfc0009cd6502260e411505d3ff254a96ce)
is positive semidefinite and
![{\displaystyle \nu }](https://wikimedia.org/api/rest_v1/media/math/render/svg/c15bbbb971240cf328aba572178f091684585468)
is a Lévy measure satisfying
![{\displaystyle \int _{\mathbb {R} ^{d}\setminus \{0\}}\min(|y|^{2},1)\nu (dy)<\infty }](https://wikimedia.org/api/rest_v1/media/math/render/svg/71d872f384d97321024e1a606b11fce267135ceb)
and
![{\displaystyle 0\leq 1-\chi (s)\leq \kappa \min(s,1)}](https://wikimedia.org/api/rest_v1/media/math/render/svg/2b83c5ab7c68d960323602e4d6da1ed28cf2a882)
for some
![{\displaystyle \kappa >0}](https://wikimedia.org/api/rest_v1/media/math/render/svg/f2d8b6e1180bd22cf9e92b0295ede259cb80db64)
with
![{\displaystyle s\chi (s)}](https://wikimedia.org/api/rest_v1/media/math/render/svg/976c7aec156c01c851e94198eb4d85cd2e24e22f)
is bounded. If we define
![{\displaystyle \psi (\xi )=\psi (0)-il\cdot \xi +{\frac {1}{2}}\xi \cdot Q\xi +\int _{\mathbb {R} ^{d}\setminus \{0\}}(1-e^{iy\cdot \xi }+i\xi \cdot y\chi (|y|))\nu (dy)}](https://wikimedia.org/api/rest_v1/media/math/render/svg/c840b0dd741c075ba00d4ce68de5edcf6caa397f)
for
![{\displaystyle \psi (0)\geq 0}](https://wikimedia.org/api/rest_v1/media/math/render/svg/b2cb1dc86ede6f0d9b532e62de8c17c90b650df1)
then the generator can be written as
![{\displaystyle Af(x)=-\int e^{ix\cdot \xi }\psi (\xi ){\hat {f}}(\xi )d\xi }](https://wikimedia.org/api/rest_v1/media/math/render/svg/cb436a085bd62dc5bc3a1fc216fe03dd671c5d38)
where
![{\displaystyle {\hat {f}}}](https://wikimedia.org/api/rest_v1/media/math/render/svg/14ce989fd75da938ec6f95a0cdb71037b23a11cb)
denotes the Fourier transform. So the generator of a Lévy process (or semigroup) is a Fourier multiplier operator with symbol
![{\displaystyle -\psi }](https://wikimedia.org/api/rest_v1/media/math/render/svg/db179e0c31a9ae42d7e993a683a5ca791245ec7d)
.
Stochastic differential equations driven by Lévy processes[edit]
Let
be a Lévy process with symbol
(see above). Let
be locally Lipschitz and bounded. The solution of the SDE
exists for each deterministic initial condition
and yields a Feller process with symbol
Note that in general, the solution of an SDE driven by a Feller process which is not Lévy might fail to be Feller or even Markovian.
As a simple example consider
with a Brownian motion driving noise. If we assume
are Lipschitz and of linear growth, then for each deterministic initial condition there exists a unique solution, which is Feller with symbol
![{\displaystyle q(x,\xi )=-il(x)\cdot \xi +{\frac {1}{2}}\xi Q(x)\xi .}](https://wikimedia.org/api/rest_v1/media/math/render/svg/503c7a27586260c181a1705f58e6da2b4749044b)
Mean first passage time[edit]
The mean first passage time
satisfies
. This can be used to calculate, for example, the time it takes for a Brownian motion particle in a box to hit the boundary of the box, or the time it takes for a Brownian motion particle in a potential well to escape the well. Under certain assumptions, the escape time satisfies the Arrhenius equation.[2]
Generators of some common processes[edit]
For finite-state continuous time Markov chains the generator may be expressed as a transition rate matrix.
The general n-dimensional diffusion process
has generator
![{\displaystyle {\mathcal {A}}f=(\nabla f)^{T}\mu +tr((\nabla ^{2}f)D)}](https://wikimedia.org/api/rest_v1/media/math/render/svg/63856e6fe1c2912cdfb523b5ebdf3784affada01)
where
![{\displaystyle D={\frac {1}{2}}\sigma \sigma ^{T}}](https://wikimedia.org/api/rest_v1/media/math/render/svg/30da081a3dfe5a819523b84bb0b83549bb1d81e7)
is the diffusion matrix,
![{\displaystyle \nabla ^{2}f}](https://wikimedia.org/api/rest_v1/media/math/render/svg/f41018628f3a828a558d68269557e7cd3ec7342e)
is the
Hessian of the function
![{\displaystyle f}](https://wikimedia.org/api/rest_v1/media/math/render/svg/132e57acb643253e7810ee9702d9581f159a1c61)
, and
![{\displaystyle tr}](https://wikimedia.org/api/rest_v1/media/math/render/svg/e049e3abf8645c35b94e206366b2ee4e0a5da76b)
is the
matrix trace. Its adjoint operator is
[2]![{\displaystyle {\mathcal {A}}^{*}f=-\sum _{i}\partial _{i}(f\mu _{i})+\sum _{ij}\partial _{ij}(fD_{ij})}](https://wikimedia.org/api/rest_v1/media/math/render/svg/bfc6eb07ecf4021b071003eb9cd6faf5cb878e88)
The following are commonly used special cases for the general n-dimensional diffusion process.
- Standard Brownian motion on
, which satisfies the stochastic differential equation
, has generator
, where
denotes the Laplace operator.
- The two-dimensional process
satisfying: ![{\displaystyle \mathrm {d} Y_{t}={\mathrm {d} t \choose \mathrm {d} B_{t}}}](https://wikimedia.org/api/rest_v1/media/math/render/svg/7043945393fc6b8119b1b445ee92699bb6f82f4c)
where
is a one-dimensional Brownian motion, can be thought of as the graph of that Brownian motion, and has generator: ![{\displaystyle {\mathcal {A}}f(t,x)={\frac {\partial f}{\partial t}}(t,x)+{\frac {1}{2}}{\frac {\partial ^{2}f}{\partial x^{2}}}(t,x)}](https://wikimedia.org/api/rest_v1/media/math/render/svg/14aac939f11bf58d642975e6d084e11865cb6533)
- The Ornstein–Uhlenbeck process on
, which satisfies the stochastic differential equation
, has generator: ![{\displaystyle {\mathcal {A}}f(x)=\theta (\mu -x)f'(x)+{\frac {\sigma ^{2}}{2}}f''(x)}](https://wikimedia.org/api/rest_v1/media/math/render/svg/52855f80e5303a96fc183dbb90352f63c5127d58)
- Similarly, the graph of the Ornstein–Uhlenbeck process has generator:
![{\displaystyle {\mathcal {A}}f(t,x)={\frac {\partial f}{\partial t}}(t,x)+\theta (\mu -x){\frac {\partial f}{\partial x}}(t,x)+{\frac {\sigma ^{2}}{2}}{\frac {\partial ^{2}f}{\partial x^{2}}}(t,x)}](https://wikimedia.org/api/rest_v1/media/math/render/svg/0b77520ae7d666c3e3637871a63efb9865d672fc)
- A geometric Brownian motion on
, which satisfies the stochastic differential equation
, has generator: ![{\displaystyle {\mathcal {A}}f(x)=rxf'(x)+{\frac {1}{2}}\alpha ^{2}x^{2}f''(x)}](https://wikimedia.org/api/rest_v1/media/math/render/svg/65d9ba7020f1f17272c8de0a2ee37b782f724fc7)
See also[edit]
References[edit]