Choose one of two regimes once, with equal probability: a source of independent and identically distributed random variables with one-bit probability , or one with one-bit probability . Keep the chosen regime forever. The resulting source is a strictly stationary process, because shifting a block does not change either component law. It is not an ergodic stationary process: its limiting empirical frequency identifies the regime.
The source's information rate isConditioning on the single regime bit changes any block's information entropy by at most one bit, so dividing by the block length proves this rate. However, the limiting realized information content per symbol is approximately in the first regime and one in the second. The alternative component's likelihood ratio decays exponentially along a typical path from the chosen component, at a rate given by the positive relative entropy between the two Bernoulli distributions. The mixture therefore has two possible pathwise limits, not one limit equal to the averaged information rate.
Stationary distribution Created 2026-09-24 Updated 2026-10-14
A stationary distribution is a probability distribution preserved by a Markov chain's evolution. For a Discrete-time Markov chain in the row convention,Thus is a normalized nonnegative left eigenvector of the transition matrix for eigenvalue one. Starting with distributed as makes every have that same probability distribution. For a time-homogeneous Markov chain, the path probabilities are also unchanged by shifting all observation times together: the process has strict stationarity. This does not mean its individual paths are constant. See Hao Wu's MIT lecture 2, stationary distributions.
For the sunny/rainy transition matrix, write . The sunny component of gives , so andMoreover,These equal stationary flows verify detailed balance and show that the Markov chain is reversible. The diagonal detailed balance equations hold automatically. Only of the stationary days are rainy, but of the days following a rainy day are rainy: stationarity permits dependence.
A finite Markov chain that is an irreducible Markov chain has a unique stationary distribution. If it is also an aperiodic Markov chain, for every initial probability distribution. The sunny/rainy example satisfies both conditions: both directions are possible and each state has a positive holding probability of a Markov chain. See Hao Wu's MIT lecture 4, convergence theorem. For an infinite state space, the additional positive recurrence hypothesis in the countable-state Markov chain convergence theorem matters.
Existence of a stationary distribution alone does not give this convergence. Consider the deterministic two-cycleHere , but starting in the first state gives alternating probability distributions . Its period of a state in a Markov chain is two. The running fraction of visits to either state nevertheless tends to , illustrating the Markov chain ergodic theorem. Starting with instead keeps the one-time probability distributions stationary even though every path alternates.
Stationary distribution of a graph random walk Created 2026-09-24 Updated 2026-11-05
On a finite connected loopless undirected graph with at least two graph vertices, the stationary distribution of a graph random walk for the simple random walk isMore generally, for positive symmetric edge electrical conductances with no killing, replace the vertex degree by and by . The detailed balance identity proves stationarity directly.
The stationary distribution is uniform precisely when the connected unweighted graph is regular. For a star with leaves, the center has stationary mass and each leaf has stationary mass . Starting at the center, the probability distribution alternates between center and leaves, so it does not converge to stationarity without an aperiodicity mechanism such as a positive holding probability. For a disconnected graph, different closed components admit different stationary mixtures; isolated graph vertices require an explicit transition convention.
Stationary distribution of a regenerative countdown chain Created 2026-10-06 Updated 2026-11-05
For the geometric-jump countdown Markov chain, regeneration at zero givesA cycle visits positive state precisely when its jump reaches at least , with Probability . Dividing this expected occupation by the mean cycle length proves the formula. Mean recurrence times are , and the limiting Probability away from zero is .
Stationary jump traffic at a single contact Created 2026-10-25 Updated 2026-11-05
Use the master-equation convention that is the transition rate from to . In a stationary law , the mean directed transition count per unit time isThe departure-state occupation multiplies the departure-to-destination rate. The reversed rate index would describe a different transition. A physical event transporting signed charge in the forward direction has mean charge current .[ref]
Stochastic thermodynamics, fluctuation theorems, and molecular machines by Udo Seifert. arXiv:1205.4176v1, Section 6.1.1, equations (110)–(111): rate convention and probability current
For a closed finite chain whose complete transition graph is a tree, every stationary edge current is zero. Removing one edge separates the states into two sets; stationarity makes total probability flow across that cut vanish, and the removed edge is its only crossing. Both directed counts can nevertheless be positive. On a three-state path with symmetric adjacent rates , the stationary law is uniform and the two edge traffic rates are and in each direction.
Different physical transition mechanisms between the same states should be kept as separate edges of a multigraph when their currents matter. Summing their rates produces an aggregated edge. Opposing mechanism currents can cancel in that aggregate, so zero aggregate current does not force every mechanism's current to vanish.
Stationary position of a transposed distinguished book Created 2026-10-31 Updated 2026-11-05
Place books in a row and distinguish one of them. At each step select the distinguished book with Probability and each of the other books with Probability . Swap the selected book with its left neighbour; selecting the leftmost book leaves the ordering unchanged.
The distinguished book's position is a birth-death chain. From an interior position it moves left with Probability , moves right with Probability when the book immediately to its right is selected, and otherwise holds. Missing outward moves at the endpoints become holding. Detailed balance therefore gives for . With ,This finite Truncated geometric distribution is valid even when : both numerator and denominator then have the same negative sign. The uniform case is , when every book is selected equally often. Large selection Probability for the distinguished book biases it toward the left end. The strict range of ensures positive motion in both directions and avoids division by zero in the ratio.
Stationary-reset perturbation of a Markov chain Created 2026-09-28 Updated 2026-11-05
Let be a transition matrix with Stationary distribution , and let have every row equal to . For , the stationary-reset perturbationfollows the original transition with Probability and independently resamples from stationarity with Probability . Since and ,The total variation identity is exact: subtracting stationarity multiplies every signed row difference by the same nonnegative factor. Once the first reset occurs the distribution remains stationary, so the factor also has a direct no-reset Probability interpretation.
This can destroy pre-cutoff. Take the dimension- Shift-refresh chain on a hypercube, whose distance is for and zero for . Use reset Probability for fixed and . For , , while at the distance is zero. A positive fraction of the mixing scale therefore already causes a nonvanishing decrease from one. Any scale capable of making the later distance tend to zero must be at least order , and its earlier fixed positive multiple cannot have distance tending to one. Thus this perturbed family lacks pre-cutoff even though the original has a sharp cutoff.
Stationary weights of a reflected birth-death chain Created 2026-10-25 Updated 2026-11-05
Consider an irreducible Markov chain on the nonnegative integers with up-step probabilities and down-step probabilities for , reflected at zero. The displayed weights satisfy detailed balance. Every stationary distribution is proportional to them: the stationary equation at zero sets the edge flow to zero, and the remaining stationary equations make the flow constant along all edges. Hence positive recurrence is equivalent to , and then . This also allows holding probabilities of a Markov chain.
Stationary workload of an exponential-payload queue Created 2026-10-14 Updated 2026-11-05
With packet arrival rate lambda, independent exponential payloads of mean one and byte-service rate C, let . The number N of unfinished packets has law . Conditional on N=n, the remaining payloads are n independent unit-mean exponentials, including the in-service residual by memorylessness. Summing their Laplace transforms gives . Thus W has mass at zero and, conditional on positivity, is exponential with rate . The probability of at least r headers is instead for integer r.
Statistical hypothesis test Created 2026-10-14 Updated 2026-11-05
A statistical hypothesis test is a decision rule calibrated to distinguish a null hypothesis from a specified alternative hypothesis. A randomized statistical test gives a rejection probability for possible data . Its power function of a statistical test is , and a level- test satisfiesThe supremum is the size of a statistical test. Controlling it limits type I error; a useful test also has high statistical power against scientifically relevant alternatives.
For ten independent Bernoulli trials with success probability , test against by rejecting for at least nine successes. The upper-tail probability increases with , so the largest null rejection probability occurs at and equalsIt is therefore a conservative level-0.05 test. Rejecting at eight successes would have size . This illustrates why discrete calibration need not attain the advertised level exactly. A p-value is a null-tail probability for the observation, not a probability that the null hypothesis is true. The decision also does not measure the size or practical importance of an effect.
Statistical inference Created 2026-09-24 Updated 2026-10-13
Statistical inference uses observed data and a statistical model to learn about unknown features of the process that generated them. In a parametric statistical model, write : the probability distribution depends on an unknown statistical parameter . An estimand specifies the target, such as a mean or a difference of means. A statistic is a function of the observations; an estimator is a statistic used to estimate the target. Uppercase denotes the random data before observation, while lowercase denotes their realized values.
Three questions lead to different outputs. Statistical estimation asks which target value the observations suggest. A confidence interval asks for a range produced by a procedure with calibrated coverage probability. A hypothesis test asks whether the observations depart sufficiently from a specified null hypothesis to enter its rejection region. These outputs complement one another: an estimated effect size describes magnitude, while a standard error describes the sampling distribution of an estimator. A p-value describes a tail probability under the null hypothesis.
The appeal of statistical inference is that it turns variable observations into reproducible uncertainty statements. In the normal mean confidence interval with known variance below, a sample mean of has standard error : both the estimated shift and its precision matter. The calibration depends on the statistical model, including independence and the specified probability distribution. Increasing sample size improves precision under those assumptions; it does not repair a systematically misleading sampling procedure. Bayesian inference provides another framework, using a prior distribution and a likelihood function to obtain a posterior distribution.
Statistical learning Created 2026-09-24 Updated 2026-10-14
Statistical learning constructs predictive rules from data and studies their prediction error, generalization, and computational fitting. In supervised learning, training data contain inputs and a response variable : regression analysis predicts a numerical response, while classification in statistical learning predicts a label. Unsupervised learning instead seeks structure without supplied responses, as in cluster analysis and principal component analysis. These tasks connect learning to probability and mathematical optimization. [ref]
https://web.stanford.edu/~hastie/ISLR2/Slides/Ch2_Statistical_Learning.pdf. James, Witten, Hastie and Tibshirani, ISLR chapter 2 slides
For a fitted rule and loss function , the predictive goal iswhere is a fresh observation, independent of the training data , from the intended target probability distribution. This prediction error fixes the fitted rule and averages over future observations; averaging again over training data gives the learning procedure's average performance. Its practical beauty is that a rule compresses past observations into useful future predictions. Its central difficulty is that excellent training error need not imply good generalization: cross-validation supports model selection, and a reserved test set assesses the resulting choice.
Statistical model Created 2026-09-24 Updated 2026-11-05
A statistical model is a specified family of probability distributions for the observations. A statistical parameter often indexes it as , where is the parameter space. The choice states which patterns of variation are considered possible and supplies a basis for statistical inference; it does not guarantee that any member exactly describes the data-generating mechanism.
For example, suppose binary observations are independent and identically distributed with a common unknown success probability . The Bernoulli distribution model has and likelihood functionWith and seven successes, this becomes and is maximized at . The binary observation type alone does not supply independence or a common : those are additional model assumptions. A regression model can instead let the response law depend on observed covariates, while a mixed model represents a specified grouping structure through random effects.
Statistical physics Created 2026-09-24 Updated 2026-10-13
Statistical physics explains collective behavior using probability distributions over microscopic possibilities. Its central surprise is that incomplete microscopic knowledge can still produce precise predictions for internal energy, entropy and heat capacity. The bridge is statistical mechanics: specify the available microstates, assign their probabilities, then compute the expected values of measured quantities.
A two-level system already displays the idea. Warming it makes its higher energy more likely; its heat capacity first rises, then falls as the two probabilities approach equality. With two distinguishable units, merely counting microstates explains why the middle energy has twice the state degeneracy. These small examples expose the machinery before the much larger sums needed for a photon gas or an Ising model.
The appeal is economical: the same microscopic energies and probabilities explain both an average reading and its variance. A measurement's spread therefore becomes useful physical information rather than an inconvenience to discard.
Statistical power Created 2026-09-24 Updated 2026-11-05
The statistical power of a test at an alternative parameter value is the probability that the test rejects its Null hypothesis at that value.
Statistical significance Created 2026-10-14 Updated 2026-11-05
A result is statistically significant at a specified significance level when its valid p-value is at most and the corresponding null hypothesis is rejected. Significance does not measure effect size, establish practical importance, or give the probability that the null is true. Software stars encode thresholds for -values; they do not supply an additional statistic.
Statistical test Created 2026-10-05 Updated 2026-11-05
A statistical test is a measurable rule for deciding whether to reject a null hypothesis. Allowing randomisation, it is represented by a function giving the conditional rejection probability after observing . A deterministic test takes only values zero and one. Under a simple null, is its size of a statistical test; under an alternative, is its statistical power. Randomisation on a likelihood ratio equality set allows a Neyman-Pearson lemma threshold test to attain a prescribed size even for discrete data.
Stellar collision cross-section Created 2026-10-20 Updated 2026-11-05
For two approximately spherical stars approaching with speed , contact at separation has gravitationally focused cross section . This measures direct stellar encounters, not the probability of a collective encounter between their host galaxies.
A phase bounded by mass-dependent lifetimes and occupies , clipped to the population age range. Its boundaries can be mapped into uniform probability coordinates to evaluate evolutionary-state fractions.
Stellar noise Created 2026-10-20 Updated 2026-11-05
Stellar variability that contaminates a planetary radial-velocity or photometric signal. Rotating active regions, changes in convective line shifts, oscillations and granulation can produce apparent velocities. A common spectral shift, line-shape diagnostics, activity indicators and simultaneous photometry provide different tests; none is an automatic numerical probability.
