Metropolis Algorithm¶

In statistics and statistical physics, the Metropolis–Hastings algorithm is a Markov chain Monte Carlo (MCMC) method for obtaining a sequence of random samples from a probability distribution from which direct sampling is difficult.

Version: v1 · 2026-09-28 · History
Domain-specific #: 10708
Domain group: Natural Sciences
Origin domain: Physics
Subdomains: Statistical Physics, Computational Physics, Markov Chain Monte Carlo → Physics

Core Idea¶

Metropolis Algorithm is treated here as the recurring Markov-chain Monte Carlo identity summarized by this source-grounded definition: In statistics and statistical physics, the Metropolis–Hastings algorithm is a Markov chain Monte Carlo (MCMC) method for obtaining a sequence of random samples from a probability distribution from which direct sampling is difficult.

In statistics and statistical physics, the Metropolis–Hastings algorithm is a Markov chain Monte Carlo (MCMC) method for obtaining a sequence of random samples from a probability distribution from which direct sampling is difficult. New samples are added to the sequence in two steps: first a new sample is proposed based on the previous sample, then the proposed sample is either added to the sequence or rejected depending on the value of the probability distribution at that point. The resulting sequence can be used to approximate the distribution (e.g. to generate a histogram) or to compute an integral (e.g. an expected value).

Metropolis–Hastings and other MCMC algorithms are generally used for sampling from multi-dimensional distributions, especially when the number of dimensions is high. For single-dimensional distributions, there are usually other methods (e.g. adaptive rejection sampling) that can directly return independent samples from the distribution, and these are free from the problem of autocorrelated samples that is inherent in MCMC methods. A common choice for g(x\mid y) is a Gaussian distribution centered at y , so that points closer to y are more likely to be visited next, making the sequence of samples into a Gaussian random walk.

For Metropolis Algorithm, the abstraction is narrower than the article's general subject matter: a positive case must preserve In statistics and statistical physics, the Metropolis–Hastings algorithm is a Markov chain Monte Carlo (MCMC) method for obtaining a sequence of random samples from a probability distribution from which direct sampling is difficult. Retaining only the name, a familiar example, or a downstream effect is insufficient. The specialist roles and tests remain anchored in Markov-chain Monte Carlo, which is why this identity is domain-specific rather than prime.

Structural Signature¶

Sig role-phrases:

Defining carrier — This is guaranteed by ergodicity of the Markov process, which requires that every state must (1) be aperiodic—the system does not return to the same state at fixed intervals; and (2) be positive recurrent—the expected number of steps for returning to the same state is finite.
Constitutive relation — A Markov process is uniquely defined by its transition probabilities P(x' \mid x) , the probability of transitioning from any given state x to any other given state x'.
Operating condition — The algorithm is named in part for Nicholas Metropolis, the first coauthor of a 1953 paper, entitled Equation of State Calculations by Fast Computing Machines, with Arianna W.
Recognition evidence — The generalized method was eventually identified by both names, although the first use of the term "Metropolis-Hastings algorithm" is unclear.
Admissible variation — Further historical clarification is made by Gubernatis in a 2005 journal article recounting the 50^th anniversary conference.
Characteristic consequence — This contradicts an account by Edward Teller, who states in his memoirs that the five authors of the 1953 article worked together for "days (and nights)".
Failure boundary — In contrast, the detailed account by Rosenbluth credits Teller with a crucial but early suggestion to "take advantage of statistical mechanics and take ensemble averages instead of following detailed kinematics".

What It Is Not¶

Not the whole field of Markov-chain Monte Carlo. The node requires the specific identity stated by In statistics and statistical physics, the Metropolis–Hastings algorithm is a Markov chain Monte Carlo (MCMC) method for obtaining a sequence of random samples from a probability distribution from which direct sampling is difficult.
Not an over-broad reading. However, prior to 2003 there was no detailed account of the algorithm's development.
Not an over-broad reading. However, if we attempt to move to a less probable point, we will sometimes reject the move, and the larger the relative drop in probability, the more likely we are to reject the new point.
Not an over-broad reading. Even though over the long term they do correctly follow P(x) , a set of nearby samples will be correlated with each other and not correctly reflect the distribution.
Not automatically Monte Carlo Simulation. Retrieval proximity does not establish equivalence; the two identities must be compared by carrier, operation, and failure boundary.

Scope of Application¶

Metropolis Algorithm applies literally inside Markov-chain Monte Carlo wherever the source-defined carrier and relation can be established. Its documented habitats include:

Description. The method used to propose new candidates is characterized by the probability distribution g(x\mid y) (sometimes written Q(x\mid y) ) of a new proposed sample x given the previous sample y.
Description. For the purpose of illustration, the Metropolis algorithm, a special case of the Metropolis–Hastings algorithm where the proposal function is symmetric, is described below.
Description. On the other hand, most simple rejection sampling methods suffer from the "curse of dimensionality", where the probability of rejection increases exponentially as a function of the number of dimensions.
Description. As a result, MCMC methods are often the methods of choice for producing samples from hierarchical Bayesian models and other high-dimensional statistical models used nowadays in many disciplines.
Description. Various algorithms can be used to choose these individual samples, depending on the exact form of the multivariate distribution: some possibilities are the adaptive rejection sampling methods, the adaptive rejection Metropolis sampling algorithm, a simple one-dimensional Metropolis–Hastings step, or slice sampling.
History. Metropolis, who was familiar with the computational aspects of the method, had coined the term "Monte Carlo" in an earlier article with Stanisław Ulam, and led the group in the Theoretical Division that designed and built the MANIAC I computer used in the experiments in 1952.

Outside Markov-chain Monte Carlo, the name should be retained only when these same operational conditions survive; otherwise the comparison belongs to the broader parent Monte Carlo Simulation or should be marked as analogy.

Clarity¶

A clear use of Metropolis Algorithm names the carrier, the operative relation, and the conditions under which the source treats the identity as present. The minimal definition is In statistics and statistical physics, the Metropolis–Hastings algorithm is a Markov chain Monte Carlo (MCMC) method for obtaining a sequence of random samples from a probability distribution from which direct sampling is difficult. The strongest recognition evidence in the frozen account is: The generalized method was eventually identified by both names, although the first use of the term "Metropolis-Hastings algorithm" is unclear. A report should distinguish that evidence from a proxy, consequence, or common implementation. It should also state the qualification However, prior to 2003 there was no detailed account of the algorithm's development. so that a reader can reproduce the classification rather than infer it from topical resemblance.

Manages Complexity¶

Metropolis Algorithm compresses multiple Markov-chain Monte Carlo details into a stable diagnostic relation. The source shows both the central mechanism—a Markov process is uniquely defined by its transition probabilities P(x' \mid x) , the probability of transitioning from any given state x to any other given state x' .—and the practical consequence—this contradicts an account by Edward Teller, who states in his memoirs that the five authors of the 1953 article worked together for "days (and nights)". This compression makes cases comparable while leaving parameters, conventions, exceptions, and evidential quality explicit. It is lossy by design: local history and implementation details may be omitted only when they do not alter the defining relation.

Abstract Reasoning¶

Type the carrier. Identify the Markov-chain Monte Carlo entities to which the claim applies.
State the relation. Use the source-grounded identity: In statistics and statistical physics, the Metropolis–Hastings algorithm is a Markov chain Monte Carlo (MCMC) method for obtaining a sequence of random samples from a probability distribution from which direct sampling is difficult.
Check operation and conditions. The algorithm is named in part for Nicholas Metropolis, the first coauthor of a 1953 paper, entitled Equation of State Calculations by Fast Computing Machines, with Arianna W.
Demand recognition evidence. The generalized method was eventually identified by both names, although the first use of the term "Metropolis-Hastings algorithm" is unclear.
Test variation. Change an implementation or setting while preserving further historical clarification is made by Gubernatis in a 2005 journal article recounting the 50^th anniversary conference.
Run the collapse test. Remove the defining operation; if the label still seems equally apt, only a topic or correlate was retained.
Reduce cautiously. When the specialist conditions cannot be carried, route the residual comparison to Monte Carlo Simulation.

Knowledge Transfer¶

Within the home domain. Knowledge about Metropolis Algorithm transfers literally when a new case preserves the same carrier type, relation, and recognition test. The method used to propose new candidates is characterized by the probability distribution g(x\mid y) (sometimes written Q(x\mid y) ) of a new proposed sample x given the previous sample y. For the purpose of illustration, the Metropolis algorithm, a special case of the Metropolis–Hastings algorithm where the proposal function is symmetric, is described below.

Beyond the home domain. Transfer the broader Monte Carlo Simulation relation when the Markov-chain Monte Carlo-specific differentia cannot be filled. Retain the name Metropolis Algorithm only when the same carrier, operation, and rejection conditions are present literally rather than metaphorically.

Examples¶

Canonical¶

The paper proposed the algorithm for the case of symmetrical proposal distributions, but in 1970, W.K. This case is canonical because it supplies a concrete carrier and lets the defining relation be checked rather than merely named.

Mapped back: carrier → the entities in the documented case; operation → In statistics and statistical physics, the Metropolis–Hastings algorithm is a Markov chain Monte Carlo (MCMC) method for obtaining a sequence of random samples from a probability distribution from which direct sampling is difficult; recognition evidence → The generalized method was eventually identified by both names, although the first use of the term "Metropolis-Hastings algorithm" is unclear

Applied / In Practice¶

More complicated proposal functions are also possible, such as those of Hamiltonian Monte Carlo, Langevin Monte Carlo, or preconditioned Crank–Nicolson. The applied case shows how the identity is used under a second setting or qualification while keeping the same operative relation.

Mapped back: changed setting → Description; invariant → In statistics and statistical physics, the Metropolis–Hastings algorithm is a Markov chain Monte Carlo (MCMC) method for obtaining a sequence of random samples from a probability distribution from which direct sampling is difficult; boundary → the case exits the class when however, prior to 2003 there was no detailed account of the algorithm's development

Structural Tensions¶

T1 — Stable identity versus admissible variation. However, prior to 2003 there was no detailed account of the algorithm's development. The tension matters because emphasizing only one side either dissolves the identity or overstates what the evidence and domain conventions warrant.

Diagnostic: Which changes preserve the defining relation, and which replace it?

T2 — Recognition versus proxy. However, if we attempt to move to a less probable point, we will sometimes reject the move, and the larger the relative drop in probability, the more likely we are to reject the new point. The tension matters because emphasizing only one side either dissolves the identity or overstates what the evidence and domain conventions warrant.

Diagnostic: Does the cited evidence establish the identity or only a correlated sign?

T3 — Definition versus implementation. Even though over the long term they do correctly follow P(x) , a set of nearby samples will be correlated with each other and not correctly reflect the distribution. The tension matters because emphasizing only one side either dissolves the identity or overstates what the evidence and domain conventions warrant.

Diagnostic: Is the observed implementation constitutive, optional, or merely common?

T4 — Scope versus overextension. Although the Markov chain eventually converges to the desired distribution, the initial samples may follow a very different distribution, especially if the starting point is in a region of low density. The tension matters because emphasizing only one side either dissolves the identity or overstates what the evidence and domain conventions warrant.

Diagnostic: Can every claimed application fill the same typed roles without metaphor?

T5 — Transfer versus domain accent. This is guaranteed by ergodicity of the Markov process, which requires that every state must (1) be aperiodic—the system does not return to the same state at fixed intervals; and (2) be positive recurrent—the expected number of steps for returning to the same state is finite. The tension matters because emphasizing only one side either dissolves the identity or overstates what the evidence and domain conventions warrant.

Diagnostic: Does the receiving case instantiate Metropolis Algorithm literally, co-instantiate Monte Carlo Simulation, or only resemble it?

T6 — Autonomy versus reduction. A Markov process is uniquely defined by its transition probabilities P(x' \mid x) , the probability of transitioning from any given state x to any other given state x'. The tension matters because emphasizing only one side either dissolves the identity or overstates what the evidence and domain conventions warrant.

Diagnostic: What does Metropolis Algorithm distinguish that the broader parent Monte Carlo Simulation leaves together?

Structural–Framed Character¶

Metropolis Algorithm is mixed or framed-leaning. Its structural side is the repeatable organization summarized by In statistics and statistical physics, the Metropolis–Hastings algorithm is a Markov chain Monte Carlo (MCMC) method for obtaining a sequence of random samples from a probability distribution from which direct sampling is difficult. Its framed side is the Markov-chain Monte Carlo vocabulary that fixes the carrier, evidence, exceptions, and admissible transformations.

Evaluative weight: the identity can be stated descriptively even when applications carry practical stakes. Human-practice dependence: the source-grounded carrier determines whether the relation exists independently or is constituted by a practice. Institutional origin: disciplinary conventions stabilize the name and test. Vocabulary portability: The algorithm is named in part for Nicholas Metropolis, the first coauthor of a 1953 paper, entitled Equation of State Calculations by Fast Computing Machines, with Arianna W. Import versus recognition: literal transfer requires the same mechanism; shape alone is analogy.

Its portable skeleton is Monte Carlo Simulation. Its character: a recurring specialist identity whose thin organization can be abstracted, while its operational meaning remains domain-bound.

Structural Core vs. Domain Accent¶

What is skeletal. In statistics and statistical physics, the Metropolis–Hastings algorithm is a Markov chain Monte Carlo (MCMC) method for obtaining a sequence of random samples from a probability distribution from which direct sampling is difficult. The reviewed portable genus is Monte Carlo Simulation; the candidate preserves that parent relation across admissible variants. The source-grounded carrier and relation are expressed by these conditions: This is guaranteed by ergodicity of the Markov process, which requires that every state must (1) be aperiodic—the system does not return to the same state at fixed intervals; and (2) be positive recurrent—the expected number of steps for returning to the same state is finite. A Markov process is uniquely defined by its transition probabilities P(x' \mid x) , the probability of transitioning from any given state x to any other given state x'. The recognition and variation tests add: The algorithm is named in part for Nicholas Metropolis, the first coauthor of a 1953 paper, entitled Equation of State Calculations by Fast Computing Machines, with Arianna W. The generalized method was eventually identified by both names, although the first use of the term "Metropolis-Hastings algorithm" is unclear.

What is domain-bound. Markov-chain Monte Carlo fixes the carrier, technical vocabulary, admissible evidence, and exceptions that distinguish Metropolis Algorithm from other Monte Carlo Simulation instances. Its documented habitat includes the condition that The method used to propose new candidates is characterized by the probability distribution g(x\mid y) (sometimes written Q(x\mid y) ) of a new proposed sample x given the previous sample y. A second source-grounded application condition is that For the purpose of illustration, the Metropolis algorithm, a special case of the Metropolis–Hastings algorithm where the proposal function is symmetric, is described below. Those details determine what the words denote, what observations warrant classification, and which apparent similarities are false positives.

Why the node remains domain-specific. Removing the Markov-chain Monte Carlo differentia leaves the parent rather than the candidate. The edge records that reduction without claiming that every topical neighbor is hierarchical. The final collapse test is source-specific: Further historical clarification is made by Gubernatis in a 2005 journal article recounting the 50^th anniversary conference. If that condition or the defining relation is absent, the case may instantiate Monte Carlo Simulation, but it is not Metropolis Algorithm.

This entry is a kind of Monte Carlo Simulation.

Immediate parent — Monte Carlo Simulation (subsumption). Metropolis Algorithm is a domain-specific kind of Monte Carlo Simulation. Metropolis Algorithm is a strict kind of Monte Carlo Simulation: In statistics and statistical physics, the Metropolis–Hastings algorithm is a Markov chain Monte Carlo (MCMC) method for obtaining a sequence of random samples from a probability distribution from which direct sampling is difficult. The parent supplies the necessary broader identity—Random sampling approximation.—while the candidate adds its domain carrier, relation, and rejection conditions.
Other nearby abstractions. Retrieval neighbors remain comparison surfaces only; no additional parent is asserted without a necessary-genus or structural-prerequisite test.

Relationships to Other Abstractions¶

Current abstraction Metropolis Algorithm Domain-specific

Parents (1) — more general patterns this builds on

Metropolis Algorithm is a kind of Monte Carlo Simulation Prime

Metropolis Algorithm is a strict kind of Monte Carlo Simulation: In statistics and statistical physics, the Metropolis–Hastings algorithm is a Markov chain Monte Carlo (MCMC) method for obtaining a sequence of random samples from a probability distribution from which direct sampling is difficult.

Hierarchy paths (4) — routes to 4 parentless roots

Metropolis Algorithm → Monte Carlo Simulation → Approximation → Representation → Abstraction

Show alternative paths (3)

Neighborhood in Abstraction Space¶

Metropolis Algorithm sits in a moderately populated region (45^th percentile for distinctiveness): it has near-neighbors but no dense thicket of look-alikes.

Family — Markov Chains & Probabilistic Computation (6 abstractions)

Nearest neighbors

Nearly completely decomposable Markov chain — 0.88
Borel right process — 0.87
Big O in probability notation — 0.86
Telescoping Markov chain — 0.86
Busy beaver — 0.86

Computed from structural-signature embeddings · 2026-10-08

Not to Be Confused With¶

Monte Carlo Simulation. The parent omits the specialist differentia. Tell: Can the case establish In statistics and statistical physics, the Metropolis–Hastings algorithm is a Markov chain Monte Carlo (MCMC) method for obtaining a sequence of random samples from a probability distribution from which direct sampling is difficult?
Monte Carlo Simulation. Random sampling approximation. Tell: Which entry's carrier, operation, and failure condition are satisfied?
Monte Carlo method in statistical mechanics. The use of stochastic sampling, commonly Markov-chain transitions, to estimate equilibrium or path-ensemble observables from high-dimensional statistical-mechanical distributions. Tell: Which entry's carrier, operation, and failure condition are satisfied?
Metropolis light transport. A global-illumination rendering method that uses Metropolis-Hastings mutations of complete light-transport paths to concentrate samples on image contributions that are difficult to find independently. Tell: Which entry's carrier, operation, and failure condition are satisfied?
A measurement, proxy, or consequence. Those may provide evidence without being the identity. Tell: Would Metropolis Algorithm remain present if the detector or downstream effect changed?
A metaphorical analogue. A similar shape outside Markov-chain Monte Carlo lacks the specialist mechanism. Tell: Do the native roles transfer literally, or only the parent Monte Carlo Simulation?

References¶

Frozen Wikipedia discovery revision: https://en.wikipedia.org/wiki/Metropolis%E2%80%93Hastings_algorithm (revision 1351697575).
Preserved source candidate: https://projecteuclid.org/journals/annals-of-statistics/volume-22/issue-4/Markov-Chains-for-Exploring-Posterior-Distributions/10.1214/aos/1176325750.full
Preserved source candidate: https://apps.dtic.mil/sti/citations/tr/ADA640705
Preserved source candidate: https://dx.doi.org/10.1109/MCS.2018.2876959
Preserved source candidate: https://www.aip.org/history-programs/niels-bohr-library/oral-histories/28636-1
Preserved source candidate: https://zenodo.org/record/1231899
Preserved source candidate: http://www.stat.columbia.edu/~gelman/research/published/theory7.ps
Preserved source candidate: https://archive.org/details/springer_10.1007-978-1-4757-4145-2
Preserved source candidate: https://www.worldscientific.com/worldscibooks/10.1142/5602

The frozen Wikipedia revision is discovery provenance. The retained source set was reviewed for identity, formal or operational relation, and scope. The encyclopedia's structural synthesis is bounded to those claims; a thin authority surface is recorded as a nonblocking source-strengthening repair rather than concealed.