Let be a probability mass function on a finite alphabet with numerical outcomes . Prescribe a mean strictly between the smallest and largest outcomes. Exponential tilting gives
with chosen so that . For any other feasible law , direct substitution yields
Relative entropy nonnegativity therefore proves optimality and uniqueness. At an extreme prescribed mean, feasible laws may have zero probabilities; a finite tilting parameter need not realize them, although a limit can.
Let be the event of membership in the first of two groups and the event of selection. Write , and , with . Bayes' theorem gives the conditional probabilities and . Their risk ratio is the displayed expression: the unknown overall selection probability cancels. Equivalently, it is the ratio of the group-membership odds after selection to those before selection.
When both selection probabilities lie strictly between zero and one, the selection odds ratio is and generally depends on . Thus compositions alone do not identify that odds ratio or provide sample sizes for a standard error. When , the first group's observed selection rate is lower; this comparison alone does not establish a causal effect of group membership.
Relative state of a bipartite vector Created 2026-10-14 Updated 2026-11-05
For a bipartite vector and a chosen orthonormal basis of , the generally unnormalized relative state on is obtained by the partial inner product with one basis vector. It gives the expansion . The squared norm of a relative state is the probability of its corresponding basis outcome when is normalized. Relative states of eigenvectors of a Choi matrix construct the columns of its Kraus operators.
Reliability diagram 2026-11-05
A reliability diagram compares predicted probabilities with observed outcome frequencies to assess probability calibration. For the positive-class version of binary classification, partition observations into bins according to their predicted probabilities . For every nonempty bin, plot
where is for a positive outcome and otherwise. Agreement places the estimated points near the diagonal . Empty bins have no observed fraction and should not be plotted as zero.[ref]
Binning can hide important structure. Consider ten cases each assigned , of which eight are positive, and ten cases each assigned , of which two are positive. Separate bins give and , revealing strong disagreement. Merge all 20 cases into one bin: the mean prediction becomes and the positive fraction is . The single point lies on the diagonal despite the two-group discrepancy.
Figure 1.
Merging bins can hide disagreement in a reliability diagram.
Original synthetic binary classification example with the same 20 predictions and outcomes in both panels. Points show positive-class frequencies; the diagram does not establish a population guarantee. Details.
Report bin definitions and counts, since a point based on ten cases has different sampling uncertainty from one based on thousands. Some classification diagrams instead plot confidence in the selected label against the fraction of correct selections. That convention and the positive-class version answer related but distinct questions, so label the axes and the meaning of the outcome explicitly.[ref]
Removal of a half-oscillator wall Created 2026-10-18 Updated 2026-11-05
Suddenly remove the wall of a half-harmonic oscillator in its ground state, leaving the same harmonic potential on the whole line. The sudden approximation keeps its wavefunction unchanged during removal. The new normalized ground state is , so
The integration is over the initially occupied half-line. The final first excited state instead has probability , since the initial half-line state equals times its positive-half restriction. This also checks that removing the wall does not simply create the full oscillator's odd state.
Renewal equation Created 2026-10-06 Updated 2026-11-05
A renewal equation decomposes a process at its first renewal. For returns of a Markov chain to a fixed state, is the first-return probability and the probability of being at that state at time . Their probability generating functions satisfy .
Renewal interval containing a fixed time Created 2026-09-24 Updated 2026-11-05
The renewal interval containing a fixed time has length
The interval is selected by a time rather than by its index. In a stationary renewal process with positive interarrival law and finite mean , its distribution is size-biased:
The renewal-reward theorem explains this: give a cycle reward equal to its length when that length lies in , and zero otherwise. The long-run fraction of time spent in such cycles is exactly the displayed ratio. A longer cycle occupies more observation time and therefore has a higher chance of containing a typical observation.
If interarrivals are one or three time units with equal probabilities, an indexed cycle has probability of being long, but an equilibrium time falls in a long cycle with probability . The expected observed interval length is , compared with indexed mean two. This is an inspection effect, not a change to the underlying interarrival law. For a renewal process freshly started at zero, the exact finite-time distribution need not already equal the equilibrium size-biased law.
Renewal representation Created 2026-10-06 Updated 2026-11-05
For positive interarrivals with law and locally bounded forcing , the unique locally bounded solution of is , where is the renewal measure. Iteration proves the identity: the residual is bounded on each compact interval by a constant times the probability that the sum of many positive interarrivals stays in that interval, which tends to zero.
For a normalized quantum state and a time-independent Hamiltonian operator , consider ideal projective measurements of the orthogonal projection at intervals . Write . A successful check leaves the post-measurement state up to global phase. Thus each successive success, conditional on earlier successes, has conditional probability , and multiplication gives the displayed formula. Equivalently,
This is the probability of the entire sequence of successful checks. It underlies the quantum Zeno effect.
Repeated singleton-removal allocation Created 2026-10-31 Updated 2026-11-05
Fix bins. In repeated uniform balls-in-bins allocation, throw each remaining ball independently into those same bins, remove every ball that is alone in its bin, and repeat with the survivors. Let be the number remaining before round .
Conditional on , a specified ball is alone with probability . Its survival probability is therefore , so linearity of expectation gives
For or the next count is zero. Here denotes the fixed bin count, not the current number of balls; confusing those two quantities changes the process.
The survival indicator random variables are generally dependent. For two initial balls and two bins, they survive together exactly when they choose the same bin, of probability each round. Hence and for . In particular, replacing the conditional formula by the same nonlinear function of would not be justified by linearity of expectation.
Repeated weak-rotation bomb-test efficiency Created 2026-10-14 Updated 2026-11-05
For an Elitzur-Vaidman bomb tester using a rotation and its inverse, let be the probability of entering the absorbing path and . Conditional on a live absorber, one trial has explosion probability , successful safe identification probability , and inconclusive probability . The inconclusive projective measurement resets the input, so summing a geometric series gives the displayed success probability after at most trials. Its limit is , approaching one half as tends to zero through positive values. Exactly zero coupling produces no information.
Repetition code Created 2026-10-03 Updated 2026-11-05
A binary repetition code represents one bit by repeating it times, using codewords and . Their Hamming distance is , and the code rate is . For odd , the majority function recovers the input whenever fewer than half the positions have flipped. Even allows a tied vote, so a decoder must specify how it handles that case.
For , the received words decode to zero; decode to one. The decoder therefore corrects every single flip. This guarantee concerns the number of flipped positions, independent of any probabilistic model.
If each position passes independently through a binary symmetric channel with crossover probability , a decoded error requires at least two flips. The binomial distribution gives
At , the uncoded error probability is , while the three-bit code's decoded error probability is . This improvement costs three transmitted bits per message bit. For , ; for , one should first exploit the known tendency to invert symbols rather than applying the same majority rule blindly.
Figure 1.
Three-bit majority decoding corrects one flip and lowers the error probability at the cost of rate
. Details.
Independent flips are an essential assumption for the displayed formula. If a single common disturbance flips all three positions together with probability , repetition provides no improvement. More efficient error-correcting codes distribute redundancy across larger message blocks; repetition is useful as a transparent example of the geometry and the reliability–rate tradeoff.
Replicating portfolio in a binomial market Created 2026-09-24 Updated 2026-10-14
Let be the initial monetary balance in the bank account, negative for borrowing. Matching successor financial payoffs gives the linear system
Subtracting determines the stock holding in the header; substitution determines . The initial cost is . For the European call option with , , strike price 100 and payoffs 20 and zero,
The debt becomes 40 at maturity: half a share minus 40 yields in the up state and in the down state. Holding these positions until maturity is self-financing. Repeating this two-equation construction at earlier nodes gives backward option pricing and the unique price in the complete market.
Figure 1.
A half-share position and a debt of 800/21 reproduce the call payoff in both states
. Details.
The discrete replication method was developed in Cox, Ross and Rubinstein's 1979 paper, whose Sections 2–3 explain borrowing, statewise matching and why physical probability drops out of relative option valuation. Their symbols for physical and pricing probabilities differ from the convention here. [ref]
Replication in experimental design Created 2026-10-07 Updated 2026-10-18
Replication in experimental design repeats a treatment on multiple independent experimental units, so that variation between those units can inform a treatment contrast and its standard error. Repeated measurements on one experimental unit can improve measurement precision without providing additional independent treatment replication. The experimental unit is identified by where the treatment is applied, rather than by how many responses are recorded.[ref]
For a constructed example, grow one plant in each of eight separate pots. Apply treatment A to four pots and treatment B to four pots, without sharing drainage water between pots. Measure the length of three leaves per pot. The pot is the experimental unit; the leaf is the observational unit. There are four treatment replicates per group, eight experimental units overall, and measurements. The three leaves in one pot are nested measurements, because they share the same plant and treatment application. The setup assumes that pots do not affect one another; simply putting them in separate containers does not establish statistical independence.
Figure 1.
Constructed two-treatment design: eight pot experimental units contain twenty-four leaf observational units
. Each outlined pot receives one treatment application and supplies three nested measurements. A pot label marks an experimental unit, not a measured response value. Details.
Randomization chooses which experimental units receive which treatments; replication supplies multiple experimental units receiving the same treatment. These solve different problems. In the balanced allocation above, choose the four A pots uniformly among the possible subsets. A particular pot has probability of receiving A. Given that one pot receives A, a second receives A with conditional probability , rather than . Thus the assignments are not independent random variables, even though the experimental units can supply independent responses conditional on those assignments in a suitable statistical model.
An actual ecology experiment illustrates why the distinction matters. In Stuart H. Hurlbert's account, twelve experimental ponds initially received zero, fifty or 450 fish, with five, four and three ponds respectively. Preparation included assessing variation between ponds before introducing fish; allocation considered both location and existing differences between ponds. The zero-fish ponds supplied the scientific control. These were repeated treatment applications to ponds, not twelve samples from one pond.[ref]
The original article followed the southern California ponds for ten months. Its readouts included zooplankton populations and total phytoplankton; it reported decreases in some zooplankton populations and increased total phytoplankton.[ref] Repeated sampling within a pond helps describe its response through time, while ponds supply treatment replication. Unequal group sizes and initial pond variation matter for the statistical model; the treatment counts alone do not establish equal variances or independent responses.
Judging probability through resemblance to a typical case or generating process. Resemblance need not incorporate prevalence or evidence reliability, so it can yield plausible-seeming but poorly calibrated probability judgments.
Resonance envelope of a Rabi transition Created 2026-10-18 Updated 2026-10-20
In the Driven two-level Rabi model, the transition probability is . Maximizing over time at each detuning gives a Lorentzian line shape, with first maximizing time and full width at half maximum in angular frequency. The amplitude parameter must use the specified Rabi angular frequency convention.
This envelope differs from a spectrum obtained with a fixed pulse duration: optimizing the time separately removes the sine-squared factor, while fixing time retains it. A finite linewidth here describes detuning-limited transfer under coherent driving, without requiring dissipative decay.
Resonance pole Created 2026-10-14 Updated 2026-11-05
A pole of the continued scattering matrix associated with outgoing boundary conditions and an unstable trapped state. A pole below the real energy axis gives time dependence , so the probability lifetime is . Bound-state poles instead lie on the positive imaginary wavenumber axis and produce square-integrable decaying spatial waves.
Resonance spin statistical factor Created 2026-10-20 Updated 2026-11-05
For unpolarized entrance particles, average over their spin orientations and sum over the isolated resonance's magnetic substates. The resulting weight multiplies the partial-wave cross section. It can exceed one because orbital angular momentum participates; it is not a standalone probability of forming a state. Polarization or several unresolved resonance spins requires more detailed channel counting.
Return level Created 2026-10-06 Updated 2026-10-14
A return level answers the inverse question to a return period: which threshold has the specified rarity? Let be the common cumulative distribution function of annual sample maxima in a model with stationarity. For a continuous, strictly increasing and , the -year return level is
It has the same units as , whereas is a count of annual blocks. Independence between years is unnecessary for this one-year quantile definition; it is needed for the geometric waiting and multi-year formulas under return period. For a discrete probability distribution, use the generalized quantile function . Then , with equality not guaranteed at jumps; an exact requested exceedance probability may be unattainable.
For a concrete standard model, choose the maxima convention of the Gumbel distribution, with location and scale :
The second formula follows by taking logarithms twice and solving . The maxima cumulative distribution function and its inverse are given in NIST: extreme value type I distribution, maximum case. With the deliberately chosen toy values and in arbitrary units, , , and , rounded to three decimals. In particular, is approximately the annual quantile at cumulative probability ; it is not a prediction of the largest observation in the next 100 years. Under independence, the probability that all those annual sample maxima stay below the exact is .
The time block matters. If a toy year contains independent daily blocks, each with exceedance probability at one fixed threshold, its annual exceedance probability is , giving annual return period . The reciprocal counts daily blocks. Dependence among daily exceedances requires their joint law instead; a threshold crossed on several successive days still produces only one exceedance indicator for that year's sample maximum. The peaks-over-threshold method and declustering of extremes address a different choice of extreme-event sampling.