Topics (384k) Articles (398k) Users (410) Discussions (242) Comments (383) Files (13k) New article
Past exam of the mathematics course of the University of Cambridge 2002 iii Paper 39 1 i Solution by
Codex 0 Created 2026-10-03 Updated 2026-10-14
Work with a nonsingular multivariate normal distribution, so is positive-definite. The claimed finite maximum-likelihood estimates also require the centered sample covariance matrix to be nonsingular; the exceptional case is discussed below.
Independence gives, up to a constant, the log-likelihoodLet and . The centered sum is zero, and expansion therefore givesFor every fixed positive-definite matrix , the extra quadratic term in the log-likelihood is , with unique maximum at .
Assume is positive-definite. Maximizing over now means minimizing . Put , another positive-definite matrix. ThenIf are its eigenvalues, the variable part is . The scalar inequality , with equality only at , shows that this is minimized uniquely at , or . ThusThe divisor is because this is maximum likelihood estimation, not the unbiased covariance estimate with divisor .
The nonsingularity condition matters. The centered scatter has rank at most . If is singular, choose , retain positive covariance eigenvalues on its range and let a covariance eigenvalue on its null space tend to zero. The quadratic likelihood term remains finite, while . This is singular covariance and nonexistence of a Gaussian maximum likelihood estimate: there is no finite maximizer over nonsingular . Under a nonsingular Gaussian population, is positive definite almost surely when ; for it is necessarily singular. The asserted estimate and the inverse in the next part have their usual meaning under .
Past exam of the mathematics course of the University of Cambridge 2002 iii Paper 38 6 iii Solution by
Codex 0 Created 2026-10-03 Updated 2026-10-14
An ancillary statistic has a sampling distribution independent of the unknown parameter. It describes a realized aspect of the sample configuration without supplying a marginal likelihood for the parameter. Its observed value can nevertheless be crucial to the precision of inference from the rest of the data. In a transitive transformation model, invariance gives a common distribution under every parameter value, so maximal invariants provide natural ancillaries, such as the standardized configuration in a location-scale family.
Conditional inference uses the sampling distribution of an estimator or sufficient statistic conditional on the observed ancillary. With joint coordinates , the factorizationshows that the conditional likelihood and the joint likelihood differ only by a parameter-free data factor. Conditional reference distributions, confidence intervals and p-values can still differ markedly from unconditional ones, because they adapt to the realized configuration. For a continuous ancillary one uses a regular conditional distribution or conditional density; dividing by the zero probability of a single ancillary value is not a valid definition.
An exact example makes this adaptation visible. Let independent observations be uniform on , with . Put , , and . Specifying the minimum and maximum contributes choices of observations, while each remaining observation must lie between them. HenceThe transformation , has unit absolute Jacobian determinant. Its support is , . Integrating over giveswhich is independent of , proving that the range is ancillary. Dividing the joint density by this marginal density givesThe centered conditional pivot is uniform, sohas conditional coverage at every , and hence unconditional coverage as well. A large realized range leaves little room for translating the interval and gives more precise inference. This is range-conditioned uniform location inference. The range and center are not independent: the conditional spread depends on the range, even though the range is ancillary.
Completeness can give independence in other models. Suppose is a complete sufficient statistic and is ancillary. For any bounded measurable , sufficiency makes a function independent of , while ancillarity makes independent of . Since for every , completeness implies almost surely. Taking all bounded indicator functions proves independence of and . This is the content of Basu's theorem, with the proof displaying precisely where sufficiency and completeness are used. Without completeness the conclusion fails: in the uniform location example the minimum and maximum are sufficient, and contain the nondegenerate ancillary range.
There are practical qualifications. Ancillaries need not be unique, and some models supply only approximate ancillary coordinates; one must specify the conditioning surface and the accuracy intended. Conditioning on a statistic sufficient for a nuisance parameter is a different device: it may remove that nuisance even though the statistic is not ancillary for the full parameter. Conditioning indiscriminately can discard useful information. In higher-order likelihood inference, the ancillary coordinates specify how the fitted data are varied, entering both the p* density and the sample derivatives in the modified profile likelihood. Their role is therefore substantive, rather than merely a label for an extra statistic.
Past exam of the mathematics course of the University of Cambridge 2002 iii Paper 38 6 ii Solution by
Codex 0 Created 2026-10-03 Updated 2026-10-14
Partition a regular parameter as , with interest and nuisance . Orthogonal statistical parameters have zero cross-block of expected Fisher information,This may hold at one parameter value or throughout a parametrization. It is expected-information orthogonality, and does not generally force the observed information cross-block to vanish in each sample.
First, the asymptotic covariance of the joint maximum-likelihood estimator is the inverse information. Orthogonality makes this matrix block diagonal, giving first-order asymptotic independence of interest and nuisance fits. The efficient information with unknown nuisance isFor orthogonal parameters it equals . Thus estimating the nuisance creates no first-order information loss relative to knowing it, at the orthogonal point. This is a local asymptotic statement, not a general finite-sample independence theorem.
Second, orthogonality stabilizes constrained nuisance estimates. Differentiating givesWhen the expected cross-information vanishes, regular sampling fluctuations give and . The derivative is therefore locally. Changing the interest parameter by changes its constrained nuisance fit by only , smaller than the ordinary nuisance estimation error. This reduces sensitivity to how the nuisance is fitted and simplifies higher-order likelihood adjustments.
For scalar interest, local orthogonal nuisance reparametrization can often be constructed explicitly. Keep and write the original nuisance as . The transformed interest score is . Its covariance with the new nuisance score vanishes whenWith smooth nonsingular nuisance information, this ordinary differential equation has a local solution for suitable initial nuisance coordinates. For several interest coordinates the corresponding equations need compatibility conditions; global orthogonality is not automatic.
The normal model gives a transparent example. For independent observations with mean and variance , direct score expectations giveThe cross-information is zero because the centered normal third moment is zero. When fitting at fixed ,so the change is indeed quadratic in a local change of the mean.
Orthogonality is also useful for removing nuisance effects beyond first order. Take variance as interest and mean as nuisance. Put . The profile log-likelihood is up to a constant. The constrained mean is for all , so its sample-coordinate Jacobian determinant is one, and . The information adjustment givesThis agrees with the exact likelihood based on : estimating the mean has used one degree of freedom. In this setting the Cox-Reid adjusted profile likelihood has the same form. The exact distribution follows by projecting the centered normal vector onto the -dimensional subspace orthogonal to the constant vector; an orthonormal basis there supplies independent standard normal coordinates whose squares sum to .
The example shows both the benefit and the limits. Orthogonality organizes first-order precision and nuisance stability; it does not eliminate all higher-order nuisance effects. Nor does it identify an ancillary statistic or justify dropping the sample-coordinate factor in every modified profile likelihood. In the exponential regression calculation that factor cancels the information adjustment, so an information-only correction would give a different answer.
Past exam of the mathematics course of the University of Cambridge 2002 iii Paper 38 6 i Solution by
Codex 0 Created 2026-10-03 Updated 2026-10-14
Saddlepoint approximation methods retain the nonquadratic shape of a distribution through its cumulant-generating function. Suppose are independent and identically distributed, is finite on an open interval containing zero, and . For an interior target mean , solve the saddlepoint equation . Introduce exponential tilting byUnder the tilted distribution the mean and variance are and . For the product sample the Radon-Nikodym derivative is . Consequently the exact density relation for the sample mean isA local central limit approximation to the tilted density at its own mean gives . HenceFor the sum , the equivalent expression is , with . Under the usual smooth nonlattice density conditions this approximation has relative error of order for a fixed interior target. In the tilted Edgeworth expansion the order odd term vanishes at the mean. The exponent retains the Legendre-transform rate , rather than a quadratic Taylor approximation to that rate, explaining the useful tail and skewness behavior. Normalization is not automatic, and accuracy statements need their regularity and range restrictions.
For a normal distribution, and . Substitution recovers exactly the normal density of the sample mean. For an exponential distribution of rate , , and . The approximation becomesIt has exactly the gamma density shape. Replacing its leading constant by the exact normalizing constant makes it exact, illustrating why exactness of shape and exactness of the unnormalized approximation are separate questions.
For a vector sample mean, tilt by , solve , and use the local multivariate normal density at the tilted mean. This givesJoint and marginal saddlepoint densities can also be divided to approximate conditional densities, for example after conditioning on a nuisance sufficient statistic. The p* approximation uses likelihood and fitted-information coordinates to obtain a closely related approximation to a maximum-likelihood estimator's density, usually conditional on an ancillary statistic.
Tail probabilities need a corresponding integration approximation. The Lugannani-Rice saddlepoint tail approximation usesand givesHere are the standard normal distribution functions. The upper-tail correction has the opposite sign: . Inverting a cumulative distribution function introduces a pole in addition to the density's saddlepoint; treating both supplies the correction term, which is why simply integrating a normal approximation at the untilted mean is less accurate. At the apparent singularities cancel. With and , expansions give and , so . For the normal distribution the correction is zero everywhere.
These methods require care for lattice distributions, where summation and lattice corrections replace the density inversion, and near boundaries or singular covariance matrices. A distribution with no moment-generating function in a neighborhood of zero does not satisfy the starting assumptions. A successful saddlepoint approximation is therefore a structured refinement of regular asymptotic inference, not a formula guaranteed uniformly in every tail or for every model.
Past exam of the mathematics course of the University of Cambridge 2002 iii Paper 37 6 v Solution by
Codex 0 Created 2026-10-03 Updated 2026-10-14
For full siblings in an outbred pedigree, Mendelian IBD sharing of full siblings gives prior sharing probabilities for . Let be the event that both siblings are affected. Bayes' rule weights those probabilities by the pair penetrances, giving immediatelyTo obtain the one-shared-copy probability, assign each allele a risk multiplier , equal to with probability and with probability . Then and . With , write the siblings' penetrances as and , where the shared multiplier and the two unshared multipliers are independent. Conditional independence of the disease outcomes givesThe total pair probability is therefore , and the three ascertained sharing probabilities areIn particular,The one-IBD proportion is below one half for a polymorphic locus with a genuine penetrance effect. Equality holds when , namely when or ; there is then no variation in the allele risk multiplier. The inequality follows from the different ascertainment weights for zero, one and two shared copies, rather than from applying the unascertained sibling probabilities to affected pairs. This is sibling IBD ascertainment under multiplicative penetrance. All affected-pair conditional probabilities presume ; if disease never occurs, conditioning on affected pairs is undefined.
Past exam of the mathematics course of the University of Cambridge 2002 iii Paper 37 6 iv Solution by
Codex 0 Created 2026-10-03 Updated 2026-10-14
For binary traits, is the indicator that both individuals are affected. Hence the pair probability is its expectation, or equivalently the covariance plus the product of individual means. Using the preceding results,These statements retain the same conditional-independence and common-population assumptions as the multiplicative diallelic penetrance covariance calculation.
Past exam of the mathematics course of the University of Cambridge 2002 iii Paper 37 6 iii Solution by
Codex 0 Created 2026-10-03 Updated 2026-10-14
Use the usual genetic model in which disease outcomes are conditionally independent given the two individuals' genotypes, and unshared founder alleles are independently drawn from the same outbred population. Let denote their number of allele copies shared by identity by descent. Conditional independence then givesWhen , both individuals have the same genotype . Its distribution remains the population genotype distribution under Hardy-Weinberg equilibrium, so their individual trait means are andThereforeWhen , all four unshared copies are independent, making independent population genotypes. Thus andThe residual-independence assumption is necessary. Marginal penetrances alone do not determine joint disease probabilities. For example, with , setting both traits equal to one shared Bernoulli variable of probability preserves every marginal penetrance, but gives covariance even with no shared alleles. Shared environmental disease determinants or population structure therefore require an expanded model. The displayed answers are the intended locus-only covariance results, not a consequence of the penetrance table without that usual assumption.
Past exam of the mathematics course of the University of Cambridge 2002 iii Paper 37 6 ii Solution by
Codex 0 Created 2026-10-03 Updated 2026-10-14
Since the binary trait has marginal Bernoulli distribution with probability , its total variance isWithin a specified genotype , the conditional variance is . The within-genotype contribution to total variance is its average over the population:Another Hardy-Weinberg equilibrium average givesHenceBy the law of total variance, the remaining between-genotype component is the genetic variance of penetrance, namelyIt is nonnegative, as a variance must be. It also follows from . The single-genotype variance and its population average are different quantities; both have been specified.
Past exam of the mathematics course of the University of Cambridge 2002 iii Paper 37 6 i Solution by
Codex 0 Created 2026-10-03 Updated 2026-10-14
Let count copies of allele . Under Hardy-Weinberg equilibrium, has a binomial distribution with parameters , and the penetrance is . ConsequentlyThus the overall penetrance isThe multiplicative diallelic penetrance model requires , , , and each occupied genotype's value to be at most one. These are probabilities, not unrestricted relative-risk scores.
Past exam of the mathematics course of the University of Cambridge 2002 iii Paper 37 5 iii Solution by
Codex 0 Created 2026-10-03 Updated 2026-10-14
Let the initial genotype frequencies be arbitrary for , with . Mendelian segregation makes the allele frequency among gametesIn random mating, the two parents are selected independently. If is a parent's transmission probability for allele , then independence gives . Thus the two independently united gametes have allele probabilities , even though the two alleles within an original parent need not have been independent. The offspring frequencies are thereforeNo initial condition was used. This proves one-generation random-mating equilibrium. The usual assumptions are a common allele pool in the two sexes, random union of gametes, ordinary Mendelian segregation, and no intervening mutation or genotype-dependent survival selection. Unequal male and female allele frequencies, for example, would instead give for , not generally the square of a single common frequency.
Past exam of the mathematics course of the University of Cambridge 2002 iii Paper 37 5 ii Solution by
Codex 0 Created 2026-10-03 Updated 2026-10-14
Treat the two parents as unordered, so three possible genotypes give six mating types. A homozygous parent transmits its allele with probability one; a heterozygous parent transmits either allele with probability one half by Mendelian segregation. Independent parental transmissions give the following complete offspring distributions:Each row sums to one. In particular, heterozygote-by-heterozygote mating can produce all three offspring genotypes, whereas the two opposite homozygotes produce only heterozygotes. These conditional offspring probabilities do not require Hardy-Weinberg equilibrium of the parents.
Past exam of the mathematics course of the University of Cambridge 2002 iii Paper 37 5 i Solution by
Codex 0 Created 2026-10-03 Updated 2026-10-14
Under Hardy-Weinberg equilibrium, the two parental allele copies are sampled independently from the same pool. Multiplying their frequencies, and adding the two orderings for a heterozygote, givesTheir sum is . The genotypes are unordered even though the paternal and maternal transmissions may be distinguished in the calculation.
Past exam of the mathematics course of the University of Cambridge 2002 iii Paper 37 4 v Solution by
Codex 0 Created 2026-10-03 Updated 2026-10-14
The source plot's sharp rise in the final year crosses the chosen control boundary. Trigger a prompt independent investigation of the mortality signal. First verify records, deaths, follow-up and expected-count calculations, then review case mix, changes in referral or recording, and plausible clinical explanations. The response should include appropriate clinical governance and safeguards while the cause is assessed; a statistical signal is not, by itself, proof of misconduct or a particular cause.
The stated null average run length of years makes a signal unusual under the calibrated null, while the alternative value of years indicates sensitivity to a sustained doubling under the specified model. Neither number gives the posterior probability that a doubling has occurred, and is not automatically the false-alarm probability for this particular ten-year history. Likewise the threshold is not a standalone fixed-test significance level.
The earlier value says only that, at that point, no recent accumulated positive log-likelihood ratio survived the resets. It is not evidence that the earlier death rate was certainly normal, and does not cancel the later signal. The zero reset is deliberately part of the change-detection design; the 1994–1996 increments can independently justify an alarm. The full earlier clinical history remains relevant to investigation, even though a negative earlier accumulated score is not carried into the new CUSUM segment.
Past exam of the mathematics course of the University of Cambridge 2002 iii Paper 37 4 iv Solution by
Codex 0 Created 2026-10-03 Updated 2026-10-14
Iterating the CUSUM recursion shows thatThus it searches all possible starting times for a recent stretch of evidence favoring an increase. If adding an increment makes the current sum negative, that accumulated evidence favors the null; retaining it would force a later real deterioration first to repay a deficit accumulated during healthy years. Starting a new candidate segment gives zero instead. Resetting at zero makes detection responsive to a change with unknown onset. Negative values would be permissible for a fixed-start signed log-likelihood ratio, but that is a different statistic from this one-sided tabular CUSUM.
Past exam of the mathematics course of the University of Cambridge 2002 iii Paper 37 4 iii Solution by
Codex 0 Created 2026-10-03 Updated 2026-10-14
Let the annual count be , with null mean and alternative mean . Dividing the two probability mass functions for the Poisson distributions cancels and the common power of :Consequently the score for the Poisson likelihood-ratio CUSUM isAt , this becomesUnder the stated Poisson model this identity is exact, rather than an additional approximation. The approximation in a practical application concerns the mortality model and the use of an estimated expected count.
Past exam of the mathematics course of the University of Cambridge 2002 iii Paper 37 4 ii Solution by
Codex 0 Created 2026-10-03 Updated 2026-10-14
For a specified change, use the log-likelihood-ratio score. A log-likelihood ratio accumulates comparable evidence for the alternative hypothesis against the null hypothesis, includes the changing expected number of deaths, and yields the standard CUSUM designed to detect that alternative. Under the null its expected increment is negative; under the target increase it is positive.
By contrast, raw has zero null drift and does not specify which increase is being sought. Its noise scale changes with expected exposure, since a Poisson count with mean has variance . A reset-at-zero chart with this zero-drift score can wander upwards through chance variation, and its threshold has no direct likelihood-evidence interpretation. An chart can be redesigned with an allowance and appropriately calibrated limits, but unadjusted is not the same as the Poisson likelihood-ratio CUSUM.
Past exam of the mathematics course of the University of Cambridge 2002 iii Paper 37 4 i Solution by
Codex 0 Created 2026-10-03 Updated 2026-10-14
For a CUSUM threshold , define the first signalling timeThe average run length is under a specified distribution of the observations and specified initial state, here . The in-control value is computed under the null hypothesis; a large value means false signals occur infrequently. The out-of-control value is computed under a specified alternative hypothesis; a small value means rapid detection when that alternative holds from monitoring's start. If a change occurs later, its detection delay depends also on the chart state at change time, so the zero-state alternative average run length is not every possible delay.
A fixed-horizon statistical hypothesis test is characterized by Type I error probability under its null and Type II error probability under a specified alternative. Continuous monitoring repeatedly offers opportunities to signal: over an indefinitely long run its probability of eventually signalling can be one even under the null. The average run length quantifies waiting time rather than that eventual probability. It replaces the single-test emphasis on a false-rejection probability with a false-signal timescale, and compares detection speed rather than only a fixed-horizon miss probability. However, a mean run length alone does not determine either finite-horizon error probability; those require the run-length distribution.
Past exam of the mathematics course of the University of Cambridge 2002 iii Paper 37 3 iv Solution by
Codex 0 Created 2026-10-03 Updated 2026-10-14
Surveillance should be calibrated to the prevalence of concern, , rather than only to the higher rate used in the preceding calculation. With representative independent tests,Thus 50,000 tests give only about 22% detection probability at the concerning rate; a negative survey is quite likely, with probability about . It costs million. Even tests, costing million, give about detection probability at .
For a target detection probability, solveThe cost is about million. The zero-event binomial upper confidence bound gives the same perspective: after zero positives in tests, the one-sided upper bound isroughly one in , well above the concerning rate. A 50,000-animal survey is insufficient to rule out that prevalence with high confidence. A practical design should specify its desired detection probability, diagnostic sensitivity, representativeness and dependence between sampled animals. Targeting the higher-risk scrapie-positive stratum may improve efficiency if testing that stratum is feasible and its contribution to the slaughter population is estimated; its result then needs the appropriate population weighting. A cost alone does not establish adequate surveillance.
Past exam of the mathematics course of the University of Cambridge 2002 iii Paper 37 3 iii Solution by
Codex 0 Created 2026-10-03 Updated 2026-10-14
Using the upper stated scrapie-test-positive fraction, its value among slaughtered sheep is . The stipulated conditional BSE positivity then gives marginal test-positive probabilitybecause the other stratum contributes zero. If the sample consists of independently tested, representative animals, the count has a binomial distribution with parameters . ThereforeThe two results areThe Poisson distribution approximation uses mean counts and , respectively. If fewer than 2000 animals belong to the test-positive stratum, these probabilities of finding nothing are higher; the calculation at 2000 is not a guaranteed detection probability.
There is also a sampling-design distinction. If exactly positive animals in a fixed annual population of are known and the sample is without replacement, use the hypergeometric distribution instead:The given expected prevalence does not assert an exactly fixed , so the independent prevalence model is the natural interpretation of the requested calculation.
Past exam of the mathematics course of the University of Cambridge 2002 iii Paper 37 3 ii Solution by
Codex 0 Created 2026-10-03 Updated 2026-10-14
Two distinct issues are:
- Exposure measurement and comparability. Infectious-dose estimates can have substantial error, and consumption categories may differ in infectivity or processing. If the assumed cohort exposure ratio is wrong, the expected allocation used in the goodness-of-fit test is wrong. The calendar distribution within the aggregated exposure period also matters for how many onsets have matured by the observation date.
- Ascertainment of outcomes. Clinical recognition, diagnostic testing, reporting and competing mortality may differ between cohorts. Unequal detection probabilities can alter the observed onset ratio even with equal biological susceptibility and incubation periods.
Thus the comparison must account for uncertainty in both its exposure denominator and its observed-disease numerator; a very small sampling p-value does not remove those model uncertainties.
Pinned article: Introduction to the OurBigBook Project
Welcome to the OurBigBook Project! Our goal is to create the perfect publishing platform for STEM subjects, and get university-level students to write the best free STEM tutorials ever.
Everyone is welcome to create an account and play with the site: ourbigbook.com/go/register. We belive that students themselves can write amazing tutorials, but teachers are welcome too. You can write about anything you want, it doesn't have to be STEM or even educational. Silly test content is very welcome and you won't be penalized in any way. Just keep it legal!
Intro to OurBigBook
. Source. We have two killer features:
- topics: topics group articles by different users with the same title, e.g. here is the topic for the "Fundamental Theorem of Calculus" ourbigbook.com/go/topic/fundamental-theorem-of-calculusArticles of different users are sorted by upvote within each article page. This feature is a bit like:
- a Wikipedia where each user can have their own version of each article
- a Q&A website like Stack Overflow, where multiple people can give their views on a given topic, and the best ones are sorted by upvote. Except you don't need to wait for someone to ask first, and any topic goes, no matter how narrow or broad
This feature makes it possible for readers to find better explanations of any topic created by other writers. And it allows writers to create an explanation in a place that readers might actually find it.Figure 1. Screenshot of the "Derivative" topic page. View it live at: ourbigbook.com/go/topic/derivativeVideo 2. OurBigBook Web topics demo. Source. - local editing: you can store all your personal knowledge base content locally in a plaintext markup format that can be edited locally and published either:This way you can be sure that even if OurBigBook.com were to go down one day (which we have no plans to do as it is quite cheap to host!), your content will still be perfectly readable as a static site.
- to OurBigBook.com to get awesome multi-user features like topics and likes
- as HTML files to a static website, which you can host yourself for free on many external providers like GitHub Pages, and remain in full control
Figure 2. You can publish local OurBigBook lightweight markup files to either OurBigBook.com or as a static website.Figure 3. Visual Studio Code extension installation.Figure 5. . You can also edit articles on the Web editor without installing anything locally. Video 3. Edit locally and publish demo. Source. This shows editing OurBigBook Markup and publishing it using the Visual Studio Code extension. - Infinitely deep tables of contents:
All our software is open source and hosted at: github.com/ourbigbook/ourbigbook
Further documentation can be found at: docs.ourbigbook.com
Feel free to reach our to us for any help or suggestions: docs.ourbigbook.com/#contact





