Base rate 2026-10-25
A class prevalence applicable before considering the current case-specific evidence. Under random sampling from the relevant reference population, it sets the prior class probability used in Bayes' theorem.
When the prior odds of a real effect are , a study's Type I error is , and its statistical power is , a positive report updates those odds to . This follows directly from Bayes' theorem. A small false-positive rate does not itself make the effect's posterior probability high: the prior odds and power also matter. The posterior exceeds one half exactly when .
Bayes factor Created 2026-09-28 Updated 2026-10-18
A Bayes factor measures how data change the relative support for two specified statistical models. Write for their Bayesian model evidence and define the direction explicitly:
For positive finite , a value above one favors over ; a value below one favors over . Reversing the comparison gives , the opposite direction used in some formulas below.
For mutually exclusive, exhaustive with positive prior probabilities, divide their two Bayes' theorem formulas. Their common data normalizer cancels, leaving
The Bayes factor therefore multiplies prior odds into posterior odds. It is evidence conditional on the specified statistical models and their within-model prior distributions; it is not . Model prior probabilities supply a separate input.[ref]
For an original stipulated example, consider ten independent and identically distributed Bernoulli trials, with heads as success. Compare with , both fixed before seeing the data. Take to be one particular ordered sequence containing eight heads and two tails. The likelihood function under either point model is , so
If , the prior odds for against are . Hence the posterior odds are , and
Equal model prior probabilities would instead give . The same Bayes factor can thus accompany different posterior probabilities. These statements condition on the two-model comparison; they do not establish that either statistical model describes a real coin.
Figure 1.
Evidence across heads counts and its multiplication of prior odds
. Original calculated Bernoulli trials example, with ten trials and fixed point models , . Left: the Bayes factor for each possible heads count; the highlighted eight-head result favors . Right: at that fixed result, the posterior odds equal times the prior odds. The marked prior becomes posterior probability . No empirical observations are plotted. Details.
Using the heads count instead of the ordered sequence changes the sample space. The binomial distribution assigns it mass . The same binomial coefficient multiplies both models' Bayesian model evidence and cancels in their Bayes factor. Mixing an ordered-sequence numerator with a count denominator would introduce an erroneous factor of .
Fix the candidate statistical models, their parameter prior distributions, and model prior probabilities before using the current data for this update. The Bayes factor is useful because it separates evidence from model prior odds, but it still depends on the parameter prior distributions. An improper prior has an arbitrary multiplicative normalization: multiplying its kernel by multiplies its formal marginal likelihood by , even if the within-model posterior distribution is normalizable. Without a separately justified convention, the ordinary Bayes factor is then undefined up to arbitrary constants.[ref]
Bayesian box search problem Created 2026-10-03 Updated 2026-10-14
A hidden object lies in box with posterior probability . Searching box costs and, conditional on the object being there, detects it with probability . After an unsuccessful search of box , Bayes' theorem updates the probabilities to
Inference about whether sensory signals have a common source or separate sources. Bayes' theorem compares hypotheses through . A common-source hypothesis can permit cue combination; separate-source hypotheses can preserve distinct estimates. Reliability weighting conditional on a common source is therefore not the entire causal-inference problem.
Bayesian confirmation 2026-10-24
Evidence confirms relative to background when it increases its conditional probability. With and positive evidence probability, Bayes' theorem gives
Indeed, writing , and gives the posterior probability ; comparison with reduces to . Thus scientific confirmation depends on the alternatives and background, not just on a statement's logical form.
Bayesian network structure score Created 2026-10-05 Updated 2026-10-14
For observations and graph , Bayes' theorem gives . Here is the graph prior distribution and the local distribution statistical parameters. The integral is Bayesian model evidence; it averages over nuisance parameters with a proper statistical parameter prior distribution. For independent complete observations, the node-factorized likelihood function and independence of local statistical parameter prior distributions make the evidence factor over nodes; suitable conjugate priors can make the local integrals analytic. With missing node observations, integrating out unobserved values can couple the local parameters, so independence of local prior distributions alone does not guarantee this factorization. Proper priors and coherent hyperparameters matter for comparing different graphs.
Bayesian posterior Created 2026-09-24 Updated 2026-10-14
The Bayesian posterior is the conditional probability distribution of an unknown statistical parameter after observing data. Suppose a continuous prior distribution has prior density relative to Lebesgue measure , and is the sampling probability mass function or probability density function of the observed data in the statistical model. Bayes' theorem gives
The Bayesian model evidence is the normalization constant making the posterior density integrate to one. The strict positive, finite condition is essential: a zero or infinite normalizer does not define this posterior distribution, even if the unnormalized expression can be written down. Other reference measures give the corresponding density formula; discrete parameter values replace the integral by a sum. The posterior propriety article gives examples where normalization fails.
With fixed, is a likelihood function of . Its normalization as a sampling distribution is over possible data, not over parameter values. In particular, a likelihood function is not a probability density function for and need not integrate to one over . The prior density supplies the probability weighting, and the Bayesian model evidence normalizes their product. This distinction prevents a tempting mistake: reversing the conditioning merely by exchanging the words “data” and “parameter”.
For eight heads among ten conditionally independent Bernoulli trials, take the observed data to be the heads count . Its binomial distribution likelihood function and the prior density give
Using the Euler beta function , the Bayesian model evidence and normalized posterior density are
Thus the posterior distribution is . If the data are instead one particular ordered sequence with eight heads and two tails, its likelihood function omits the factor 45 and its Bayesian model evidence becomes . The factor cancels in Bayes' theorem, leaving the same posterior distribution; these two data descriptions have different sampling probabilities but identical information about here.
Figure 1.
Beta(2,2) prior, peak-normalized Bernoulli likelihood, and Beta(10,4) posterior after eight heads and two tails
. Details.
The prior density and posterior density use the left axis and each integrate to one. The dashed likelihood function uses the right axis and is divided by its maximum solely to compare its shape; it is not a parameter probability density function. The likelihood function peaks at , while the posterior mode is and the posterior mean is . Multiplying by the prior density shifts the preferred values toward one half while preserving a range of possibilities.
The resulting posterior variance is , compared with prior variance . In this example the update reduces uncertainty. The posterior predictive probability of heads on the next conditionally independent toss is , and the credible interval above reports a 95% range. These are summaries of the same posterior distribution, rather than competing estimates of one density peak.
Bayesian statistics Created 2026-09-24 Updated 2026-10-14
Bayesian statistics expresses uncertainty about an unknown statistical parameter through a probability distribution, then uses Bayes' theorem to update it after observing data. The prior distribution describes uncertainty before the current observations; the likelihood function describes how the statistical model generates those observations at each parameter value; the posterior distribution describes uncertainty conditional on the observations and that model. A physical parameter can have one fixed value while its value remains uncertain to us. Assigning it a probability distribution represents that uncertainty under the chosen model, rather than requiring the physical parameter to change randomly between measurements.
For example, suppose ten Bernoulli trials are independent conditional on an unknown heads probability , with a beta distribution prior . Eight heads and two tails give the posterior distribution by Beta-binomial conjugacy. Its posterior mean is , between the prior mean and observed binomial proportion . The Bayesian posterior below computes the whole update, and the credible interval below summarizes its remaining uncertainty. The conjugate update is explained in MIT 18.05, Reading 15: beta and normal conjugate priors.
The useful feature is that one posterior distribution supports several questions: a posterior mean estimates a value, a credible interval reports a range, and a posterior predictive distribution predicts a future observation while averaging over parameter uncertainty. With conditionally independent new data, today's posterior distribution can become tomorrow's prior distribution. These conclusions remain conditional on the statistical model and prior distribution; a precise update does not establish that either assumption fits the experiment.
A historical starting point is Thomas Bayes's essay in the 1763 volume of the Royal Society's Philosophical Transactions, communicated after his death by Richard Price. Its opening problem asks for uncertainty about an unknown event probability given observed successes and failures: the same direction of reasoning illustrated by the coin example. See Bayes and Price, An Essay towards solving a Problem in the Doctrine of Chances, opening letter and problem.
Bayesian update after a projective measurement Created 2026-10-25 Updated 2026-11-05
For candidate density matrices with prior probabilities , a projective measurement has outcome likelihood . Bayes' theorem gives the displayed posterior whenever the denominator is positive. An outcome certifies one candidate only when every other candidate has zero likelihood there. If all candidates have the same outcome likelihood, observing it leaves their prior probabilities unchanged.
For with an independent equiprobable drift , the likelihood ratio between the two drifts is . Multiplying the normal distribution densities of successive increments proves this ratio for every observation partition; continuity then gives the full observed-path posterior. The Bayes' theorem gives . Thus the observed drift is . The process is an observed-filtration martingale with quadratic variation , so the Lévy characterization of Brownian motion makes it a Brownian motion. The observed state is a scaled diffusion with hyperbolic tangent drift.
Behavior is evidence for a disposition , given a situation , to the extent that its conditional likelihood ratio differs from one. By Bayes' theorem this ratio multiplies the prior odds to give the posterior odds. If a constraint forces the same behavior regardless of the disposition, the ratio is one and the behavior leaves those odds unchanged. This does not exclude diagnostic information in other features, such as wording or voluntary choices. The argument identifies the assumptions needed to call a dispositional attribution biased.
A normative Bayesian inference model of retrospective confidence. For two exhaustive alternatives with equal prior probabilities, let be the log-likelihood ratio. Bayes' theorem gives posterior odds . Choosing the alternative with the larger posterior yields . Unequal priors add prior log odds; choosing a different action changes which posterior probability is appropriate. This calculation does not establish that an observer knows the true likelihoods.
Dermatologist 2026-11-05
A dermatologist is a physician specializing in the diagnosis and treatment of diseases affecting the skin, hair and the nail. The field combines visible patterns with information about symptoms, timing, exposures and the rest of the body. The American Academy of Dermatology describes this scope and distinguishes additional specialization in dermatopathology.[ref]
The surface is informative but does not directly reveal every cause. Similar redness or scaling can occur through different biological processes, while one process can look different at different sites or stages. A skin biopsy supplies a selected piece of tissue for microscopic examination; a dermatopathologist relates its cellular and structural findings to the clinical question.[ref] Sampling location and the question asked determine what that specimen can represent.
An original diagnostic-reasoning example shows why appearance and diagnosis are not identical. Suppose a hypothetical classification signal is positive in of affected cases and of unaffected cases, in a population where have the specified condition. Bayes' theorem gives
Thus even a signal associated with the condition can have many other explanations among positive results. These are chosen probabilities, not performance data for a skin examination, biopsy or particular diagnostic instrument.
The scientific work is to connect observations at different scales: lesion distribution, tissue architecture, cellular changes and a plausible disease mechanism. An observation that supports one mechanism need not uniquely establish it. Repeated examination can additionally reveal change over time that a single photograph cannot encode.
The profession's scope is biological and clinical, while training pathways and certification rules depend on jurisdiction. The distinction matters when interpreting an institutional role description: a particular board-certification route does not define the universal scientific meaning of dermatology.
Detection probability 2026-11-05
The detection probability is the conditional probability that a specified procedure declares a detection when the target or signal condition is present. With the detection event and the signal-present condition,
It depends on the signal, noise model, measurement conditions and decision rule. In a statistical hypothesis test, it is the statistical power at the specified alternative. Its complement is the probability of missing that signal; the false positive rate instead conditions on signal absence.[ref]
High does not by itself imply that most detections are real. Suppose, in an invented collection of10,000 independent trial conditions,100 contain a target. A procedure detecting90% of targets and falsely alarming in5% of the9,900 target-absent conditions has expected counts90 real detections and495 false alarms. Only of its alarms then correspond to targets. These are expected counts under the example model, not measured detector performance.
More generally, if and with meaning absence, Bayes' theorem gives
The order of conditioning is therefore essential. Changing a decision threshold commonly trades detection against false alarms; a receiver operating characteristic displays that tradeoff.
Discriminant analysis 2026-11-05
Discriminant analysis constructs statistical rules distinguishing labelled classes from their measured feature values. A common generative approach models the class-conditional probability distribution and applies Bayes' theorem to compare posterior class probabilities. Linear discriminant analysis uses multivariate normal distributions with a shared covariance matrix; allowing a different covariance matrix for each class gives quadratic discriminant analysis.[ref]
For two equally likely one-dimensional classes, suppose and . Their log-likelihood ratio is
Choosing the larger posterior therefore selects for , with a tie at one. Different class probabilities move this threshold. Estimating means and covariance from training data supplies a fitted rule, while independent evaluation tests whether its assumptions generalize.
Disease-ascertained transmission probability Created 2026-10-14 Updated 2026-11-05
Let be the Mendelian segregation probability, and assume the child's disease status depends on the parental genotypes only through its own genotype, with penetrance proportional to . Bayes' theorem gives the displayed conditional law. For a diallelic parent pair there are four equally likely labelled transmission outcomes. If an unordered genotype occurs in of them, , so its probability is . Dropping multiplicities would, for example, misweight a heterozygous child of two heterozygotes.
EM expectation step 2026-10-14
Hold the old statistical parameters fixed when taking the conditional expectation, while keeping the candidate statistical parameters variable:
For independent observations in a finite mixture model, is a hidden component label. Bayes' theorem gives its mixture responsibilities:
These posterior probabilities are the conditional expectations of the mixture allocation indicators; they turn hidden counts into fractional counts.[ref] They condition on a current fixed fit, rather than averaging over a posterior distribution of the statistical parameters. Taking the conditional expectation of the complete-data log-likelihood is essential: substituting a single guessed label, or merely substituting the conditional mean into a nonlinear complete-data expression, generally gives a different algorithm.
EM for zero-inflated negative binomial regression Created 2026-10-05 Updated 2026-10-14
For known size and , the expectation-maximization algorithm gives structural-zero responsibility for positive counts and for zero counts. This is Bayes' theorem. Maximizing the expected complete-data log-likelihood gives and a weighted negative binomial regression with weights . The objective separates into a Bernoulli mixing term and a weighted count term, which proves the update.
If source class has prior fraction and detection efficiency , its fraction among detected events is proportional to . Bayes' theorem normalizes these products. Class-dependent losses therefore bias the detected composition even when the incident fractions are known exactly.