{
  "id": "chapman-kolmogorov-equations",
  "title": "Chapman–Kolmogorov Equations",
  "lang": "en",
  "source_lang": "en",
  "summary": "The Chapman–Kolmogorov equations express how Markov transition probabilities compose across successive time intervals by summing or integrating over intermediate states.",
  "category": "mathematics",
  "subcategory": "probability-statistics",
  "type": "concept",
  "aliases": [
    "Chapman–Kolmogorov equation",
    "Chapman–Kolmogorov identity",
    "Chapman–Kolmogorov relation"
  ],
  "generated_at": "2026-10-04",
  "model": "pro/gpt-6.1-sol",
  "url": "https://aiwiki.page/en/chapman-kolmogorov-equations/",
  "translations": {
    "zh-Hans": {
      "url": "https://aiwiki.page/zh/chapman-kolmogorov-equations/",
      "json": "https://aiwiki.page/zh/chapman-kolmogorov-equations.json"
    }
  },
  "sections": [
    {
      "heading": "",
      "text": "The Chapman–Kolmogorov equations are composition identities for the transition probabilities of a stochastic process satisfying the Markov property. They state that a transition over a longer interval can be decomposed into transitions over two successive intervals, with all possible intermediate states accounted for. In a discrete state space, this involves a sum; in a general state space, it involves an integral against a transition probability measure. They provide the fundamental connection between local transition rules and evolution over longer times. (gordanz.github.io)"
    },
    {
      "heading": "Discrete-state formulation",
      "text": "Let \\(X_t\\) be a Markov process with a finite or countably infinite state space \\(S\\). Write\n\n\\[p_{ij}(s,t)=\\Pr(X_t=j\\mid X_s=i)\\]\n\nfor its conditional probability of moving from state \\(i\\) at time \\(s\\) to state \\(j\\) at time \\(t\\). For \\(s\\leq u\\leq t\\), the equations are\n\n\\[\\boxed{ p_{ij}(s,t)= \\sum_{k\\in S}p_{ik}(s,u)\\,p_{kj}(u,t). }\\]\n\nThe intermediate state \\(k\\) ranges over the entire state space. The identity applies whether or not the transition rules change with time. (stat.berkeley.edu)\n\nFor a time-homogeneous Markov chain, transition probabilities depend only on elapsed time. In discrete time, define\n\n\\[p_{ij}^{(n)}=\\Pr(X_n=j\\mid X_0=i).\\]\n\nThen, for nonnegative integers \\(m,n\\),\n\n\\[p_{ij}^{(m+n)} =\\sum_{k\\in S}p_{ik}^{(m)}p_{kj}^{(n)}.\\]\n\nIf \\(P^{(n)}=(p_{ij}^{(n)})\\), this becomes the matrix identity\n\n\\[P^{(m+n)}=P^{(m)}P^{(n)}.\\]\n\nConsequently, if \\(P\\) is the one-step transition matrix, then\n\n\\[P^{(n)}=P^n,\\qquad P^{(0)}=I,\\]\n\nwhere \\(I\\) is the identity matrix. The superscript in \\(P^{(n)}\\) denotes transition probabilities over \\(n\\) steps; the right-hand side \\(P^n\\) is ordinary matrix exponentiation. (gordanz.github.io)"
    },
    {
      "heading": "Probabilistic derivation",
      "text": "The derivation combines the law of total probability with the Markov property. Conditioning on the state at time \\(u\\) gives\n\n\\[\\begin{aligned} \\Pr(X_t=j\\mid X_s=i) &=\\sum_{k\\in S} \\Pr(X_t=j\\mid X_u=k,X_s=i)\\\\ &\\qquad\\qquad{}\\times \\Pr(X_u=k\\mid X_s=i). \\end{aligned}\\]\n\nThe Markov property allows the earlier condition \\(X_s=i\\) to be removed from the first factor:\n\n\\[\\Pr(X_t=j\\mid X_u=k,X_s=i) =\\Pr(X_t=j\\mid X_u=k).\\]\n\nSubstitution yields the Chapman–Kolmogorov equation. This is a statement about conditional independence: the intermediate state contains the information from the past needed to determine the future transition law. It does not assert that successive states are unconditionally independent. (gordanz.github.io)"
    },
    {
      "heading": "General state spaces and transition densities",
      "text": "On a measurable space \\((E,\\mathcal E)\\), transitions are described by a Markov kernel\n\n\\[K_{s,t}(x,A),\\]\n\nwhere \\(x\\in E\\) and \\(A\\in\\mathcal E\\). For fixed \\(x\\), this is a probability distribution over possible later states; for fixed \\(A\\), it is measurable in \\(x\\). The general identity is\n\n\\[\\boxed{ K_{s,t}(x,A) =\\int_E K_{u,t}(y,A)\\,K_{s,u}(x,dy). }\\]\n\nThe integral averages the probability of reaching \\(A\\) from each intermediate state \\(y\\), weighted by the probability of reaching that state from \\(x\\). (stat.berkeley.edu)\n\nIf the kernels have probability densities relative to a common reference measure \\(\\mu\\), the corresponding density identity is\n\n\\[p(s,x;t,z) =\\int_E p(s,x;u,y)\\,p(u,y;t,z)\\,\\mu(dy),\\]\n\nwith equality understood almost everywhere in the terminal variable unless suitably regular versions are available. The kernel formulation is more general: it also covers discrete, mixed, and singular transition laws for which an ordinary density may not exist. (stat.berkeley.edu)\n\nFor continuous states, \\(K_{s,t}(x,\\cdot)\\) is interpreted as a specified transition kernel, rather than an elementary ratio of probabilities involving the potentially zero-probability event \\(X_s=x\\). (stat.berkeley.edu)"
    },
    {
      "heading": "Semigroup interpretation",
      "text": "For a time-homogeneous process, write \\(K_t\\) for the transition kernel over duration \\(t\\). Then\n\n\\[K_{s+t}(x,A) =\\int_E K_t(y,A)\\,K_s(x,dy).\\]\n\nDefine operators on bounded measurable functions by\n\n\\[(T_tf)(x)=\\int_E f(y)\\,K_t(x,dy).\\]\n\nThese operators satisfy\n\n\\[T_{s+t}=T_sT_t,\\qquad T_0=I.\\]\n\nThus the Chapman–Kolmogorov equations express the defining composition law of a Markov semigroup. Probabilistically, \\(T_tf(x)\\) is the expected value of \\(f(X_t)\\) when the process starts at \\(x\\). Strong continuity on a chosen function space is an additional analytical condition, not part of the composition identity alone. (numerik.mi.fu-berlin.de)"
    },
    {
      "heading": "Relation to the Kolmogorov differential equations",
      "text": "For a finite-state, time-homogeneous continuous-time Markov chain, let \\(P(t)\\) be its transition matrix. The composition law is\n\n\\[P(s+t)=P(s)P(t).\\]\n\nIts infinitesimal generator is the rate matrix\n\n\\[Q=\\lim_{h\\downarrow0}\\frac{P(h)-I}{h}.\\]\n\nDifferentiating the composition identity gives the Kolmogorov forward and backward equations:\n\n\\[P'(t)=P(t)Q \\quad\\text{and}\\quad P'(t)=QP(t),\\]\n\nrespectively, with \\(P(0)=I\\). Their unique finite-state solution is the matrix exponential\n\n\\[P(t)=e^{tQ}.\\]\n\nThe forward equation isolates a short interval at the end of the transition; the backward equation isolates one at the beginning. (mpaldridge.github.io)\n\nThe Chapman–Kolmogorov equations themselves are composition identities, not differential equations. Passing to differential equations requires appropriate limiting and regularity assumptions. In infinite state spaces, interchanging limits with infinite sums requires justification, and explosion—infinitely many jumps in finite time—introduces additional complications. (metaphor.ethz.ch)"
    },
    {
      "heading": "Examples",
      "text": "For the two-state transition matrix\n\n\\[P= \\begin{pmatrix} 0.8&0.2\\\\ 0.3&0.7 \\end{pmatrix},\\]\n\ndirect multiplication gives\n\n\\[P^2= \\begin{pmatrix} 0.70&0.30\\\\ 0.45&0.55 \\end{pmatrix}.\\]\n\nFor example, the two-step probability of moving from state \\(1\\) to state \\(2\\) is\n\n\\[0.8(0.2)+0.2(0.7)=0.30.\\]\n\nThe two terms correspond to intermediate states \\(1\\) and \\(2\\).\n\nFor standard one-dimensional Brownian motion, the transition density over duration \\(t>0\\) is\n\n\\[g_t(y-x)= \\frac{1}{\\sqrt{2\\pi t}} \\exp\\!\\left[-\\frac{(y-x)^2}{2t}\\right].\\]\n\nIts Chapman–Kolmogorov identity is\n\n\\[g_{s+t}(z-x) =\\int_{\\mathbb R} g_s(y-x)\\,g_t(z-y)\\,dy.\\]\n\nThis is a convolution identity for normal distributions: independent Gaussian increments with variances \\(s\\) and \\(t\\) combine into an increment with variance \\(s+t\\). The same transition density is the fundamental solution of the heat equation with diffusion coefficient \\(1/2\\). (math.ucdavis.edu)"
    },
    {
      "heading": "Uses and scope",
      "text": "The equations propagate state distributions. With the row-vector convention, an initial distribution \\(\\mu_0\\) for a homogeneous discrete-time chain evolves according to\n\n\\[\\mu_n=\\mu_0P^n.\\]\n\nThey therefore connect a one-step model to predictions over many steps. (gordanz.github.io)\n\nThey also supply consistency conditions for constructing Markov processes from transition kernels: inserting or removing intermediate observation times must produce compatible finite-dimensional distributions. Under the appropriate state-space assumptions, these distributions can be used with the Kolmogorov extension theorem to construct a process. Such a construction does not, by itself, guarantee continuous sample paths; path regularity requires further arguments. (users.math.msu.edu)"
    }
  ],
  "links": [
    "stochastic-process",
    "markov-property",
    "conditional-probability",
    "markov-chain",
    "matrix",
    "transition-matrix",
    "identity-matrix",
    "law-of-total-probability",
    "conditional-independence",
    "measurable-space",
    "probability-distribution",
    "integral",
    "probability-density-function",
    "expected-value",
    "continuous-time-markov-chain",
    "kolmogorov-forward-and-backward-equations",
    "differential-equation",
    "brownian-motion",
    "convolution",
    "normal-distribution",
    "heat-equation"
  ],
  "sources": [
    {
      "title": "Chapter 5 Markov Chains — Lecture notes for “Introduction to Stochastic Processes”",
      "url": "https://gordanz.github.io/stochastic-book/markov-chains.html"
    },
    {
      "title": "A guide to Brownian motion and related stochastic processes",
      "url": "https://www.stat.berkeley.edu/~aldous/205B/pitman_yor_guide_bm.pdf"
    },
    {
      "title": "Lecture notes for Numerik IVc — Numerics for Stochastic Processes",
      "url": "https://numerik.mi.fu-berlin.de/wiki/WS_2012/Vorlesungen/NumerikIVc_NumerikStochastischerProzesse_Dokumente/Numerik4c_150113.pdf"
    },
    {
      "title": "Section 18 Forward and backward equations — MATH2750 Introduction to Markov Processes",
      "url": "https://mpaldridge.github.io/math2750/S18-forward-backward.html"
    },
    {
      "title": "Lecture notes on stochastic processes",
      "url": "https://metaphor.ethz.ch/x/2021/fs/401-3602-00L/sc/lecture-notes-2017.pdf"
    },
    {
      "title": "Lecture Notes on Applied Mathematics",
      "url": "https://math.ucdavis.edu/~hunter/m280_09/ch.pdf"
    }
  ],
  "license": "https://creativecommons.org/licenses/by/4.0/"
}