Fokker–Planck equation

In statistical mechanics and information theory, the Fokker–Planck equation is a partial differential equation that describes the time evolution of the probability density function of the velocity of a particle under the influence of drag forces and random forces, as in Brownian motion. The equation can be generalized to other observables as well.[1] The Fokker-Planck equation has multiple applications in information theory, graph theory, data science, finance, economics etc.

A solution to the one-dimensional Fokker–Planck equation, with both the drift and the diffusion term. In this case the initial condition is a Dirac delta function centered away from zero velocity. Over time the distribution widens due to random impulses.

It is named after Adriaan Fokker and Max Planck, who described it in 1914 and 1917.[2][3] It is also known as the Kolmogorov forward equation, after Andrey Kolmogorov, who independently discovered it in 1931.[4] When applied to particle position distributions, it is better known as the Smoluchowski equation (after Marian Smoluchowski),[5] and in this context it is equivalent to the convection–diffusion equation. When applied to particle position and momentum distributions, it is known as the Klein–Kramers equation. The case with zero diffusion is the continuity equation. The Fokker–Planck equation is obtained from the master equation through Kramers–Moyal expansion.[6]

The first consistent microscopic derivation of the Fokker–Planck equation in the single scheme of classical and quantum mechanics was performed by Nikolay Bogoliubov and Nikolay Krylov.[7][8]

One dimension edit

In one spatial dimension x, for an Itô process driven by the standard Wiener process   and described by the stochastic differential equation (SDE)

 

with drift   and diffusion coefficient  , the Fokker–Planck equation for the probability density   of the random variable   is [9]

 

Link between the Itô SDE and the Fokker–Planck equation

In the following, use  .

Define the infinitesimal generator   (the following can be found in Ref.[10]):

 

The transition probability  , the probability of going from   to  , is introduced here; the expectation can be written as

 
Now we replace in the definition of  , multiply by   and integrate over  . The limit is taken on
 
Note now that
 
which is the Chapman–Kolmogorov theorem. Changing the dummy variable   to  , one gets
 
which is a time derivative. Finally we arrive to
 
From here, the Kolmogorov backward equation can be deduced. If we instead use the adjoint operator of  ,  , defined such that
 
then we arrive to the Kolmogorov forward equation, or Fokker–Planck equation, which, simplifying the notation  , in its differential form reads
 

Remains the issue of defining explicitly  . This can be done taking the expectation from the integral form of the Itô's lemma:

 

The part that depends on   vanished because of the martingale property.

Then, for a particle subject to an Itô equation, using

 
it can be easily calculated, using integration by parts, that
 
which bring us to the Fokker–Planck equation:
 

While the Fokker–Planck equation is used with problems where the initial distribution is known, if the problem is to know the distribution at previous times, the Feynman–Kac formula can be used, which is a consequence of the Kolmogorov backward equation.

The stochastic process defined above in the Itô sense can be rewritten within the Stratonovich convention as a Stratonovich SDE:

 
It includes an added noise-induced drift term due to diffusion gradient effects if the noise is state-dependent. This convention is more often used in physical applications. Indeed, it is well known that any solution to the Stratonovich SDE is a solution to the Itô SDE.

The zero-drift equation with constant diffusion can be considered as a model of classical Brownian motion:

 

This model has discrete spectrum of solutions if the condition of fixed boundaries is added for  :

 

It has been shown[11] that in this case an analytical spectrum of solutions allows deriving a local uncertainty relation for the coordinate-velocity phase volume:

 
Here   is a minimal value of a corresponding diffusion spectrum  , while   and   represent the uncertainty of coordinate–velocity definition.

Higher dimensions edit

More generally, if

 

where   and   are N-dimensional vectors,   is an   matrix and   is an M-dimensional standard Wiener process, the probability density   for   satisfies the Fokker–Planck equation

 

with drift vector   and diffusion tensor  , i.e.

 

If instead of an Itô SDE, a Stratonovich SDE is considered,

 

the Fokker–Planck equation will read:[10]: 129 

 

Generalization edit

In general, the Fokker–Planck equations are a special case to the general Kolmogorov forward equation

 

where the linear operator   is the Hermitian adjoint to the infinitesimal generator for the Markov process.[12]

Examples edit

Wiener process edit

A standard scalar Wiener process is generated by the stochastic differential equation

 

Here the drift term is zero and the diffusion coefficient is 1/2. Thus the corresponding Fokker–Planck equation is

 

which is the simplest form of a diffusion equation. If the initial condition is  , the solution is

 

Boltzmann distribution at the thermodynamic equilibrium edit

The overdamped Langevin equation

 
gives  . The Boltzmann distribution  is an equilibrium distribution, and assuming   grows sufficiently rapidly (that is, the potential well is deep enough to confine the particle), the Boltzmann distribution is the unique equilibrium.

Ornstein–Uhlenbeck process edit

The Ornstein–Uhlenbeck process is a process defined as

 

with  . Physically, this equation can be motivated as follows: a particle of mass   with velocity   moving in a medium, e.g., a fluid, will experience a friction force which resists motion whose magnitude can be approximated as being proportional to particle's velocity   with  . Other particles in the medium will randomly kick the particle as they collide with it and this effect can be approximated by a white noise term;  . Newton's second law is written as

 

Taking   for simplicity and changing the notation as   leads to the familiar form  .

The corresponding Fokker–Planck equation is

 

The stationary solution ( ) is

 

Plasma physics edit

In plasma physics, the distribution function for a particle species  ,  , takes the place of the probability density function. The corresponding Boltzmann equation is given by

 

where the third term includes the particle acceleration due to the Lorentz force and the Fokker–Planck term at the right-hand side represents the effects of particle collisions. The quantities   and   are the average change in velocity a particle of type   experiences due to collisions with all other particle species in unit time. Expressions for these quantities are given elsewhere.[13] If collisions are ignored, the Boltzmann equation reduces to the Vlasov equation.

Smoluchowski diffusion equation edit

Consider an overdamped Brownian particle under external force  :[14]

 
where the   term is negligible (the meaning of "overdamped"). Thus, it is just  . The Fokker–Planck equation for this particle is the Smoluchowski diffusion equation:
 
Where   is the diffusion constant and  . The importance of this equation is it allows for both the inclusion of the effect of temperature on the system of particles and a spatially dependent diffusion constant.
Derivation of the Smoluchowski Equation from the Fokker–Planck Equation

Starting with the Langevin Equation of a Brownian particle in external field  , where   is the friction term,   is a fluctuating force on the particle, and   is the amplitude of the fluctuation.

 

At equilibrium the frictional force is much greater than the inertial force,  . Therefore, the Langevin equation becomes,

 

Which generates the following Fokker–Planck equation,

 

Rearranging the Fokker–Planck equation,

 

Where  . Note, the diffusion coefficient may not necessarily be spatially independent if   or   are spatially dependent.

Next, the total number of particles in any particular volume is given by,

 

Therefore, the flux of particles can be determined by taking the time derivative of the number of particles in a given volume, plugging in the Fokker–Planck equation, and then applying Gauss's Theorem.

 

 

In equilibrium, it is assumed that the flux goes to zero. Therefore, Boltzmann statistics can be applied for the probability of a particles location at equilibrium, where   is a conservative force and the probability of a particle being in a state   is given as  .

 

 

This relation is a realization of the fluctuation–dissipation theorem. Now applying   to   and using the Fluctuation-dissipation theorem,

 

Rearranging,

 

Therefore, the Fokker–Planck equation becomes the Smoluchowski equation,

 
for an arbitrary force  .

Computational considerations edit

Brownian motion follows the Langevin equation, which can be solved for many different stochastic forcings with results being averaged (canonical ensemble in molecular dynamics). However, instead of this computationally intensive approach, one can use the Fokker–Planck equation and consider the probability   of the particle having a velocity in the interval   when it starts its motion with   at time 0.

 
Brownian dynamics simulation for particles in 1-D linear potential compared with the solution of the Fokker–Planck equation

1-D linear potential example edit

Brownian dynamics in one dimension is simple.[14][15]

Theory edit

Starting with a linear potential of the form   the corresponding Smoluchowski equation becomes,

 

Where the diffusion constant,  , is constant over space and time. The boundary conditions are such that the probability vanishes at   with an initial condition of the ensemble of particles starting in the same place,  .

Defining   and   and applying the coordinate transformation,

 

With   the Smoluchowki equation becomes,

 

Which is the free diffusion equation with solution,

 

And after transforming back to the original coordinates,

 

Simulation edit

The simulation on the right was completed using a Brownian dynamics simulation.[16][17] Starting with a Langevin equation for the system,

 
where   is the friction term,   is a fluctuating force on the particle, and   is the amplitude of the fluctuation. At equilibrium the frictional force is much greater than the inertial force,  . Therefore, the Langevin equation becomes,
 

For the Brownian dynamic simulation the fluctuation force   is assumed to be Gaussian with the amplitude being dependent of the temperature of the system  . Rewriting the Langevin equation,

 
where   is the Einstein relation. The integration of this equation was done using the Euler–Maruyama method to numerically approximate the path of this Brownian particle.

Solution edit

Being a partial differential equation, the Fokker–Planck equation can be solved analytically only in special cases. A formal analogy of the Fokker–Planck equation with the Schrödinger equation allows the use of advanced operator techniques known from quantum mechanics for its solution in a number of cases. Furthermore, in the case of overdamped dynamics when the Fokker–Planck equation contains second partial derivatives with respect to all spatial variables, the equation can be written in the form of a master equation that can easily be solved numerically.[18] In many applications, one is only interested in the steady-state probability distribution  , which can be found from  . The computation of mean first passage times and splitting probabilities can be reduced to the solution of an ordinary differential equation which is intimately related to the Fokker–Planck equation.

Particular cases with known solution and inversion edit

In mathematical finance for volatility smile modeling of options via local volatility, one has the problem of deriving a diffusion coefficient   consistent with a probability density obtained from market option quotes. The problem is therefore an inversion of the Fokker–Planck equation: Given the density f(x,t) of the option underlying X deduced from the option market, one aims at finding the local volatility   consistent with f. This is an inverse problem that has been solved in general by Dupire (1994, 1997) with a non-parametric solution.[19][20] Brigo and Mercurio (2002, 2003) propose a solution in parametric form via a particular local volatility   consistent with a solution of the Fokker–Planck equation given by a mixture model.[21][22] More information is available also in Fengler (2008),[23] Gatheral (2008),[24] and Musiela and Rutkowski (2008).[25]

Fokker–Planck equation and path integral edit

Every Fokker–Planck equation is equivalent to a path integral. The path integral formulation is an excellent starting point for the application of field theory methods.[26] This is used, for instance, in critical dynamics.

A derivation of the path integral is possible in a similar way as in quantum mechanics. The derivation for a Fokker–Planck equation with one variable   is as follows. Start by inserting a delta function and then integrating by parts:

 

The  -derivatives here only act on the  -function, not on  . Integrate over a time interval  ,

 

Insert the Fourier integral

 

for the  -function,

 

This equation expresses   as functional of  . Iterating   times and performing the limit   gives a path integral with action

 

The variables   conjugate to   are called "response variables".[27]

Although formally equivalent, different problems may be solved more easily in the Fokker–Planck equation or the path integral formulation. The equilibrium distribution for instance may be obtained more directly from the Fokker–Planck equation.

See also edit

Notes and references edit

  1. ^ Leo P. Kadanoff (2000). Statistical Physics: statics, dynamics and renormalization. World Scientific. ISBN 978-981-02-3764-6.
  2. ^ Fokker, A. D. (1914). "Die mittlere Energie rotierender elektrischer Dipole im Strahlungsfeld". Ann. Phys. 348 (4. Folge 43): 810–820. Bibcode:1914AnP...348..810F. doi:10.1002/andp.19143480507.
  3. ^ Planck, M. (1917). "Über einen Satz der statistischen Dynamik und seine Erweiterung in der Quantentheorie". Sitzungsberichte der Preussischen Akademie der Wissenschaften zu Berlin. 24: 324–341.
  4. ^ Kolmogorov, Andrei (1931). "Über die analytischen Methoden in der Wahrscheinlichkeitstheorie" [On Analytical Methods in the Theory of Probability]. Mathematische Annalen (in German). 104 (1): 415–458 [pp. 448–451]. doi:10.1007/BF01457949. S2CID 119439925.
  5. ^ Dhont, J. K. G. (1996). An Introduction to Dynamics of Colloids. Elsevier. p. 183. ISBN 978-0-08-053507-4.
  6. ^ Paul, Wolfgang; Baschnagel, Jörg (2013). "A Brief Survey of the Mathematics of Probability Theory". Stochastic Processes. Springer. pp. 17–61 [esp. 33–35]. doi:10.1007/978-3-319-00327-6_2. ISBN 978-3-319-00326-9.
  7. ^ N. N. Bogolyubov Jr. and D. P. Sankovich (1994). "N. N. Bogolyubov and statistical mechanics". Russian Math. Surveys 49(5): 19—49. doi:10.1070/RM1994v049n05ABEH002419
  8. ^ N. N. Bogoliubov and N. M. Krylov (1939). Fokker–Planck equations generated in perturbation theory by a method based on the spectral properties of a perturbed Hamiltonian. Zapiski Kafedry Fiziki Akademii Nauk Ukrainian SSR 4: 81–157 (in Ukrainian).
  9. ^ Risken, H. (1996), The Fokker–Planck Equation: Methods of Solution and Applications, vol. Second Edition, Third Printing, p. 72
  10. ^ a b Öttinger, Hans Christian (1996). Stochastic Processes in Polymeric Fluids. Berlin-Heidelberg: Springer-Verlag. p. 75. ISBN 978-3-540-58353-0.
  11. ^ Kamenshchikov, S. (2014). "Clustering and Uncertainty in Perfect Chaos Systems". Journal of Chaos. 2014: 1–6. arXiv:1301.4481. doi:10.1155/2014/292096. S2CID 17719673.
  12. ^ Pavliotis, Grigorios A. (2014). Stochastic Processes and Applications : Diffusion Processes, the Fokker-Planck and Langevin Equations. Springer. pp. 38–40. doi:10.1007/978-1-4939-1323-7_2. ISBN 978-1-4939-1322-0.
  13. ^ Rosenbluth, M. N. (1957). "Fokker–Planck Equation for an Inverse-Square Force". Physical Review. 107 (1): 1–6. Bibcode:1957PhRv..107....1R. doi:10.1103/physrev.107.1.
  14. ^ a b Ioan, Kosztin (Spring 2000). "Smoluchowski Diffusion Equation". Non-Equilibrium Statistical Mechanics: Course Notes.
  15. ^ Kosztin, Ioan (Spring 2000). "The Brownian Dynamics Method Applied". Non-Equilibrium Statistical Mechanics: Course Notes.
  16. ^ Koztin, Ioan. "Brownian Dynamics". Non-Equilibrium Statistical Mechanics: Course Notes. Archived from the original on 2020-01-15. Retrieved 2020-05-18.
  17. ^ Kosztin, Ioan. "The Brownian Dynamics Method Applied". Non-Equilibrium Statistical Mechanics: Course Notes. Archived from the original on 2020-01-15. Retrieved 2020-05-18.
  18. ^ Holubec Viktor, Kroy Klaus, and Steffenoni Stefano (2019). "Physically consistent numerical solver for time-dependent Fokker–Planck equations". Phys. Rev. E. 99 (4): 032117. arXiv:1804.01285. Bibcode:2019PhRvE..99c2117H. doi:10.1103/PhysRevE.99.032117. PMID 30999402. S2CID 119203025.{{cite journal}}: CS1 maint: multiple names: authors list (link)
  19. ^ Bruno Dupire (1994) Pricing with a Smile. Risk Magazine, January, 18–20.
  20. ^ Bruno Dupire (1997) Pricing and Hedging with Smiles. Mathematics of Derivative Securities. Edited by M.A.H. Dempster and S.R. Pliska, Cambridge University Press, Cambridge, 103–111. ISBN 0-521-58424-8.
  21. ^ Brigo, D.; Mercurio, Fabio (2002). "Lognormal-Mixture Dynamics and Calibration to Market Volatility Smiles". International Journal of Theoretical and Applied Finance. 5 (4): 427–446. CiteSeerX 10.1.1.210.4165. doi:10.1142/S0219024902001511.
  22. ^ Brigo, D.; Mercurio, F.; Sartorelli, G. (2003). "Alternative asset-price dynamics and volatility smile". Quantitative Finance. 3 (3): 173–183. doi:10.1088/1469-7688/3/3/303. S2CID 154069452.
  23. ^ Fengler, M. R. (2008). Semiparametric Modeling of Implied Volatility, 2005, Springer Verlag, ISBN 978-3-540-26234-3
  24. ^ Jim Gatheral (2008). The Volatility Surface. Wiley and Sons, ISBN 978-0-471-79251-2.
  25. ^ Marek Musiela, Marek Rutkowski. Martingale Methods in Financial Modelling, 2008, 2nd Edition, Springer-Verlag, ISBN 978-3-540-20966-9.
  26. ^ Zinn-Justin, Jean (1996). Quantum field theory and critical phenomena. Oxford: Clarendon Press. ISBN 978-0-19-851882-2.
  27. ^ Janssen, H. K. (1976). "On a Lagrangean for Classical Field Dynamics and Renormalization Group Calculation of Dynamical Critical Properties". Z. Phys. B23 (4): 377–380. Bibcode:1976ZPhyB..23..377J. doi:10.1007/BF01316547. S2CID 121216943.

Further reading edit

  • Frank, Till Daniel (2005). Nonlinear Fokker–Planck Equations: Fundamentals and Applications. Springer Series in Synergetics. Springer. ISBN 3-540-21264-7.
  • Gardiner, Crispin (2009). Stochastic Methods (4th ed.). Springer. ISBN 978-3-540-70712-7.
  • Pavliotis, Grigorios A. (2014). Stochastic Processes and Applications: Diffusion Processes, the Fokker–Planck and Langevin Equations. Springer Texts in Applied Mathematics. Springer. ISBN 978-1-4939-1322-0.
  • Risken, Hannes (1996). The Fokker–Planck Equation: Methods of Solutions and Applications. Springer Series in Synergetics (2nd ed.). Springer. ISBN 3-540-61530-X.