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 2012 iii Paper 69 7 Solution by
Codex 0 Created 2026-10-03 Updated 2026-10-14
Fourier stability analysis applies most directly to linear, constant-coefficient evolution schemes on uniform infinite or periodic grids. Translation invariance makes spatial shifts diagonal in discrete Fourier modes. If a one-step scheme has , substituting gives the Fourier amplification symbolThe discrete Parseval identity identifies the discrete L2 norm with the squared norm of Fourier coefficients. Uniform stability on a fixed time interval means a mesh-independent bound on for all frequencies and . In the scalar case yields contraction. More generally gives the bound and is consistent with a continuously growing well-posed equation. A fixed modulus excess independent of the step produces growth exponential in and fails this requirement.
For systems, eigenvalue moduli alone are insufficient. The uniform power bound for matrix Fourier symbols also controls eigenvector conditioning and Jordan growth. For example has only unit eigenvalues but , unbounded as at fixed . In contrast an off-diagonal entry proportional to gives growth at most proportional to , which can be legitimate. A uniformly conditioned diagonalization is a sufficient reduction to scalar modal bounds; nonnormal matrices require direct power or energy estimates. Multilevel schemes similarly require the root condition for their frequency-dependent amplification polynomial and a uniform bound on the associated companion evolution, including limiting repeated-root cases.
The simplest parabolic example is the Forward Euler method with centered spatial diffusion. For ,is the sharp mesh-uniform contraction range. For the theta method, , giving unconditional contraction at . Implicit stability permits larger steps, but accuracy still constrains them, and the Crank-Nicolson method does not strongly damp high-frequency rough data.
For advection with , the upwind finite difference scheme givesThus is its contraction range. The Lax-Friedrichs scheme has and is contractive exactly when . These express a Courant–Friedrichs–Lewy condition: the numerical domain of dependence must encompass the physical one for convergence. That necessary condition is not sufficient by itself. Forward time with centered advection has and at nonzero modes. At a fixed nonzero Courant number it is unstable even if . Under the much stronger refinement , however, its powers remain bounded on fixed time intervals; the centered-advection refinement-dependent stability must not be confused with the usual hyperbolic-step instability.
Symbols also quantify dispersion and dissipation. For a mode , the argument of gives a numerical phase speed and gives damping. Upwind advection has leading artificial diffusion under fixed-Courant refinement; excessive damping smooths genuine structure, while phase error misplaces waves. Stability controls boundedness, not either error by itself. Smooth low-frequency Taylor expansions establish consistency; high-frequency modal bounds establish stability, so analyzing only fixed physical wavenumbers can miss grid-scale instabilities.
For a well-posed linear initial-value problem, the Lax equivalence theorem makes consistency plus a uniform stability estimate equivalent to convergence in the chosen norm. Boundaries and variable coefficients need extra care: Fourier diagonalization no longer handles the full update, and an interior or frozen-coefficient symbol does not prove boundary or global stability. Boundary energy estimates, a suitable matrix analysis or a separate boundary-stability argument must supplement it. Nonlinear schemes also require controls beyond this linear symbol test. These limitations delimit what the method proves rather than reducing its usefulness: in its proper setting, it turns an arbitrarily large grid evolution into exact and interpretable modal algebra.
Past exam of the mathematics course of the University of Cambridge 2012 iii Paper 69 6 Solution by
Codex 0 Created 2026-10-03 Updated 2026-10-14
For a discretized ordinary differential equation, error control must distinguish the local truncation error of one exact-start step from accumulated global error. For a smooth order- one-step method, these are usually and respectively, provided perturbations propagate stably. A small local defect is useful only when the underlying problem and the numerical update do not amplify it excessively.
There are several practical local error estimators. In step-doubling error estimation, compute one step and two half steps from the same starting value. If the leading method error is , the two-half-step error is , up to higher terms. Their difference therefore givesfor exact minus fine-step error. Adding that estimate is Richardson extrapolation. This costs extra evaluations but gives an independent check on a single-method implementation; nonsmooth events can invalidate the assumed expansion.
An embedded Runge-Kutta pair shares stages and uses two output weight vectors to obtain orders and . Their difference generally estimates the lower-order local error , not the higher-order error directly. Stage reuse makes this much cheaper than independent integrations. Predictor-corrector error estimation uses the appropriately calibrated difference of two formulas; raw predictor/corrector differences require their known error constants. For multistep schemes, changing the step size also changes interpolation coefficients and requires controlled step ratios and proper history updates, rather than reusing constant-step coefficients.
Component scales should reflect the requested physical accuracy. An example of absolute and relative error tolerances isPositive absolute tolerances avoid singular scaling when a component vanishes, while relative tolerances track widely differing magnitudes. Accept the step if ; otherwise reject it without committing the new state or multistep history. If the estimated defect scales as , a basic adaptive step-size controller proposesGrowth limits, a safety factor, and proportional-integral smoothing reduce erratic changes. The exponent must match the actual estimator: for a genuine order- local defect, but for the usual embedded difference. It is not always or always . A zero estimate uses a finite maximum growth factor, not division by zero.
Local acceptance does not itself certify a global error tolerance. If a dense approximation has residual and is Lipschitz continuous with constant on the relevant trajectory region, Gronwall's inequality gives the residual bound for global ODE errorThe discrete analogue propagates and sums local defects. A posteriori defect correction, repeat integrations with tighter tolerances, and adjoint-weighted estimates for a chosen final functional provide stronger global checks. A smaller local target may be needed in an unstable flow; trajectory sensitivity should not be mislabeled as a code defect.
For a stiff differential equation, absolute stability can constrain explicit steps far more tightly than accuracy alone. Use an appropriate implicit method and keep nonlinear stage/linear-solve errors below the discretization-error budget. A-stability, L-stability and B-stability address different linear damping and nonlinear contractivity properties. An error controller cannot repair a method outside its stability domain. Tolerance-driven adaptation also does not automatically preserve positivity, conservation laws or a symplectic structure; these may require a suitable method or an additional acceptance test.
Finally, event detection and dense output have their own errors. Locate an event with a consistent interpolant and account for its state sensitivity; a discontinuous right side may require restarting the integrator. Refining indefinitely eventually exposes roundoff error and conditioning limits. Reliable error control balances truncation, solution sensitivity, algebraic-solve error and floating-point error, while reporting failure to meet a target instead of endlessly shrinking a step.
Past exam of the mathematics course of the University of Cambridge 2012 iii Paper 68 3 Solution by
Codex 0 Created 2026-10-03 Updated 2026-10-07
A Schwartz function is a smooth function for which every seminormis finite. These seminorms define the Fréchet topology of the Schwartz space . The tempered distributions form its continuous linear dual . We use bilinear distribution pairing, with no complex conjugation on the test function.
Differentiation under the Fourier integral and integration by parts giveTherefore each Schwartz seminorm of is bounded by a finite sum of norms of polynomially weighted derivatives of . Insert the integrable weight to bound those norms by finitely many Schwartz seminorms. This proves that is continuous.
To justify inversion without a merely formal exchange of integrals, insert in the inverse integral. The Gaussian Fourier integral givesThe normalized Gaussian is an approximate identity, so the right side tends to ; the left side converges by dominated convergence theorem since is integrable. It follows thatReflection is continuous in Schwartz seminorms, so is continuous as well. This proves the Fourier transform isomorphism of the Schwartz space.
Define the distributional Fourier transform byIt agrees with the ordinary integral transform whenever Fubini applies. Transposing the Schwartz inverse gives an inverse on , and the same squared-transform identity holds. For the usual strong dual topology, a seminorm is for a bounded set . Since a continuous linear Schwartz map takes bounded sets to bounded sets, proves continuity, and similarly for the inverse. Continuity also holds in the weak dual topology. Thus the transform is a continuous isomorphism on tempered distributions, with the stated normalization.
For a real symmetric positive-definite matrix , there is with . The reciprocal is locally integrable in dimension three: the radial factor near zero is . At infinity, rapid decay of a Schwartz test function makes it integrable. More quantitatively,This single seminorm bound proves a tempered distribution. No principal-value extension is needed at the origin.
First calculate the isotropic transform. For and , spherical integration and the supplied sine-integral identity giveThe original functions converge in to by dominated convergence theorem against tests. The transforms are bounded by , which is locally integrable in three dimensions and integrable against Schwartz functions at infinity. HenceLet . The Jacobian is and the dual vector is . The Fourier transform of a reciprocal positive quadratic form is consequentlyThe frequency-origin value is understood distributionally; there is no additional delta term.
For the complex extension use a symmetric matrix with real symmetric and positive real part . Symmetry is natural for a quadratic form; a skew-symmetric part contributes nothing. The bound again gives a regular tempered reciprocal. Both sides of the prospective formula depend holomorphically on the matrix while its real part is positive: compact parameter sets give a common bound for pairing and differentiated integrands. Continue from the real positive matrices along . In a complex neighborhood of , imaginary gives real positive matrices, so the one-variable identity theorem supplies the equality; connected continuation along reaches .
The analytic determinant square root for accretive symmetric matrices must follow that continuation, rather than an arbitrary scalar principal root of . Explicitly, put and let its real eigenvalues be . Thenwhere each factor has positive real part. This is the determinant square root normalized positively on real positive matrices. AlsoThus the quadratic-form square root has the unambiguous branch with positive real part. The accretive complex quadratic reciprocal Fourier transform isUniform local integrability of the right side justifies its analytic continuation as a tempered distribution, not merely pointwise away from the origin.
For the particular form,Substitution gives the required normalized expressionFor every nonzero real , the radicand has positive real part, so the specified square root exists uniquely. Both sides are regular tempered distributions, despite their locally integrable singularities at zero.
Past exam of the mathematics course of the University of Cambridge 2012 iii Paper 68 2 Solution by
Codex 0 Created 2026-10-03 Updated 2026-10-07
The symbol class consists of smooth amplitudes with estimatesThere is no loss of symbol order under an derivative. A phase function is real and smooth for , positively homogeneous of degree one in , with its full differential nowhere zero there, at least on the amplitude's conic support. Homogeneous phases need not be smooth at ; that bounded-frequency region is handled separately or the amplitude is cut off near it.
Choose equal to one near zero. The oscillatory integral is defined by regularization in distributions:This is an ordinary integral when decay makes it absolutely convergent. For general order, the definition is independent of the regularizer. To see the mechanism, at large setHomogeneity and the nonvanishing full differential give on compact sets, and . Its -derivative coefficients have order and its -derivative coefficients order zero, so each formal transpose lowers the amplitude order by one. Iterating more than times makes the test-function pairing integrable, and the same estimates control regularizer derivatives. This justifies the cutoff function limit and its independence. A fixed bounded-frequency integral is smooth in .
The singular support of a distribution is the complement of the largest open set on which it is represented by a smooth function. For the support-sensitive bound one must use conic support of an oscillatory amplitude: a closed set of limiting high-frequency directions, locally in the base variable. One sufficient precise choice iswith closure taken in this product. If is outside the projection of , compactness of the unit sphere gives a neighborhood of and a positive uniform lower bound on wherever the large-frequency amplitude is supported. On that region useIts coefficients have order zero and its transpose lowers order by one. To prove regularity, first differentiate times in , raising the amplitude order by at most , and then integrate by parts more than times. The resulting integrals and their derivatives converge uniformly on compact neighborhoods. Since is arbitrary, the stationary-direction bound for singular support follows:This is the standard conic interpretation of the support restriction in the question; directions at infinity, rather than finite-frequency stationary points, determine possible singularities.
If literally means the support of the restricted function, the printed inclusion can fail. There is a slice-support obstruction for oscillatory singularities. In one dimension, take and , where is smooth and even, zero for , and one for . This is a symbol of order , and the full phase differential is nonzero for . Its regularized integral near isIndeed substitute in the tail, split at , and write below one. The first term gives , while the remainder has an even convergent power series in plus a constant; the bounded-frequency correction is smooth. Thus is in the singular support, but has empty slice support. The joint closed conic support retains this limiting base point and makes the proved theorem valid. No nonstationarity conclusion is inferred just from the amplitude vanishing at one base point.
For the wave equation, Fourier transformation in reduces the initial-value problem to , with and . For its solution iswith value at zero. Choose equal to one near zero. The low-frequency decomposition of the wave propagator iswhereThe low-frequency term is an ordinary smooth function: the sine quotient extends smoothly in at zero and all derivatives are integrable on the fixed compact frequency set. The high-frequency amplitudes belong to , vanish near zero, and the phases have nonzero full differential since . Their stationary-direction equations are . Such a direction can occur only if . The smooth low-frequency term contributes no singular support, soThis is a statement about singular support: some dimensions have a smooth nonzero tail inside the cone, so it does not claim that the distribution is concentrated only on the cone.
Past exam of the mathematics course of the University of Cambridge 2012 iii Paper 68 1 Solution by
Codex 0 Created 2026-10-03 Updated 2026-10-07
Use and , so the Fourier transform of is . Write for the homogeneous degree- part. An elliptic differential operator has for every real . By continuity on the unit sphere, . The lower-degree terms are bounded by for , whencefor sufficiently large . This proves the high-frequency lower bound for an elliptic polynomial. Degree zero just means a nonzero constant and has no derivative gain.
With the Fourier convention in Question 3, the Sobolev space isThe condition includes that the weighted transform is represented by an function. For open , the Local Sobolev space consists of for which , extended by zero, belongs to for every . We use the elementary Sobolev multiplication by a smooth cutoff fact for every real . It follows from Fourier convolution with the rapidly decreasing , the weighted inequality , and the convolution bound.
For a compactly supported distribution, continuity on test functions supported in a fixed compact neighborhood gives finite order: for some integer ,Insert a compact smooth cutoff function equal to one near the support, times . The resulting Fourier transform is smooth and satisfies . The distribution is also tempered by this same finite-order bound. ThusIndeed the square of the weighted bound is integrable exactly when . This is negative Sobolev regularity of a compactly supported distribution.
We next prove local regularity by a high-frequency reciprocal parametrix kernel, including its off-diagonal smoothing property. Choose equal to one on a ball containing all real zeros of , and put , defining it smoothly as zero in the inner ball. Differentiation of the reciprocal and the elliptic lower bound giveThe Fourier multiplier maps to by its zeroth-order bound, andThe kernel is smooth away from the origin: for , repeatedly integrate by parts using . After sufficiently many integrations, the differentiated symbol is integrable. For any desired derivative, repeat the argument with the additional polynomial . A large-radius cutoff function justifies each step and its removal uniformly on compact sets away from zero. Thus convolution by carries a compactly supported distribution to a smooth function at points separated from its support. Also carries a compactly supported distribution to a Schwartz function, because is smooth with compact support.
For a target compact set in , choose equal to one on a neighborhood of it. Then and have compact support, and the commutator is supported where derivatives of occur, away from the target. The parametrix identity givesThe first term is in since . The other two terms are smooth near the target by the proved kernel property. Since the target was arbitrary,This proof does not discard the cutoff function commutator; it places its support away from the set where regularity is sought.
For the final polynomial, select a multi-index with and a nonzero constant. The derivative-ratio hypothesis immediately gives at large frequency. In particular has no real zero there. For , its degree bound also forces . Differentiating gives products of ratios , whose total derivative order is . Hence the corresponding high-frequency reciprocal satisfiesIts multiplier maps to . Its kernel is still smooth off zero: each frequency integration by parts now lowers the order by , so more iterations may be needed, but supplies arbitrarily much decay. The same separated-support commutator identity applies without a change. The derivative-ratio Sobolev gain for a polynomial operator is thereforeSimply estimating the commutator by its differential order would lose this sharper gain; the off-diagonal kernel argument uses the full derivative-ratio hypothesis.
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





