The upper critical dimension is the dimension at which the leading nonlinear interaction is marginal under Gaussian scale transformations. Normalize the gradient term and rescale so that . A coupling then scales by
Thus
Below this dimension, the interaction is a relevant operator at the Gaussian fixed point, so neglecting long-wavelength fluctuations is not self-consistent near criticality. A direct Ginzburg criterion gives the same result: the fluctuation in a correlation volume scales as . For ordinary ordering , their ratio is proportional to . Along a tricritical trajectory , giving ratio proportional to . These ratios grow without bound below the corresponding upper critical dimension.
At , the interaction is marginal. Mean-field powers can survive, but acquire multiplicative logarithmic corrections; it is too strong to say that all mean-field exponent values necessarily fail at equality. Above , fluctuations do not change the leading mean-field critical exponents, although a dangerously irrelevant coupling can invalidate naive hyperscaling.
A tricritical point requires two independent even coefficients in the Landau free energy to vanish: , with providing stability. At , the ordinary critical line meets the tricritical three-phase line . The order parameter discontinuity along the first-order line is , which goes to zero at their meeting.
The positive sextic coefficient makes this a different leading local potential from an ordinary quartic critical point. Consequently its mean-field critical exponents, crossover behavior and upper critical dimension differ. There are two relevant even tuning directions near the tricritical point, in addition to the conjugate odd field. The root solution and labelled phase diagram display the three controls and the tricritical wings; varying can give a first-order reversal within the ordered coexistence sheet, a first-order weak/strong transition on a wing, or a continuous transition at a wing's ordinary critical edge.
Critical exponents describe leading singular powers as the reduced temperature or conjugate field approaches zero. With used for the order-parameter critical exponent to avoid confusion with inverse temperature, the definitions include
At criticality the connected correlation function behaves as . These definitions refer to the singular part and allow additional logarithms at marginal dimensions.
For an ordinary scalar Landau free energy, the equation of state is . Below the transition , giving . Differentiating the equation of state gives : it equals above and below the transition, so . The quadratic fluctuation kernel is in the symmetric phase, giving correlation length and . At zero mass the momentum-space correlation function is proportional to , so . The minimized free energy and critical equation of state give and , as calculated in the root solution.
At the tricritical point, and . Then , and is above and below the transition. The gradient kernel still gives and . Therefore
These are mean-field critical exponents. A systematic derivation incorporating fluctuations uses the renormalization group, extracts the relevant scaling eigenvalues, and differentiates the singular free-energy density, as in Question 3.
The relevant comparison is a short-range, uniaxial Ising ferromagnet and the liquid–gas critical point of water in three dimensions. Both have one real scalar order parameter: magnetization in the ferromagnet and the appropriately shifted density in the fluid. The magnetic field and a suitable pressure/chemical-potential combination provide the corresponding fields conjugate to an order parameter.
The fluid has no exact microscopic spin-reversal symmetry. Nevertheless, at the liquid–gas critical point the first three derivatives of the local thermodynamic potential can be eliminated or tuned by shifting the density and mixing the thermodynamic control variables. The leading stable expansion is then quartic, with the same long-distance renormalization-group fixed point as the three-dimensional short-range Ising model. The microscopic asymmetry produces corrections and analytic mixing of observables, rather than a different leading universality class.
Thus the scalar liquid–gas critical point and the short-range Ising ferromagnet share a universality class. This does not assert that every ferromagnet does: an isotropic vector-spin magnet or one dominated by long-range dipolar interactions can have different critical behavior. Nor does it identify every phase transition of water with its liquid–gas critical point.
Universality of critical phenomena means that many microscopic models approach the same long-distance renormalization-group fixed point. After repeated coarse-graining, irrelevant operators lose influence, whereas a few relevant operators describe deviations from criticality. Models with the same spatial dimension, order parameter symmetry, interaction range and appropriate relevant controls can therefore have the same universality class despite very different microscopic constituents.
The critical exponents, appropriately normalized scaling functions and certain amplitude ratios are universal. The transition temperature, individual amplitudes, microscopic length scale and analytic background in the free energy are generally nonuniversal. The numerical value assigned to an order parameter also depends on normalization. Landau-Ginzburg theory expresses this distinction by using a common symmetry-allowed functional form while allowing its coefficients to depend on the material. Computing the true critical exponents below the upper critical dimension requires retaining fluctuations, rather than merely minimizing that functional.
In a first-order phase transition, two distinct global minima of the Landau free energy have equal values and exchange stability. The equilibrium order parameter jumps, and a first derivative of the equilibrium free energy is discontinuous. Depending on the direction through the phase diagram, that derivative can be magnetization, entropy or another conjugate quantity; a first-order transition need not have latent heat along every chosen path.
In a continuous phase transition, the equilibrium order parameter changes continuously while the curvature at the critical minimum vanishes. For and , is stable for , but for the stable minima have . The order parameter therefore goes continuously to zero at ; the inverse magnetic susceptibility vanishes there.
For , the sextic term stabilizes the Landau free energy. The nonzero minima can become favorable while still has positive curvature. At they coexist with , so the order parameter jumps to magnitude . Finding a stationary point alone does not locate the transition: equilibrium requires comparing the values of all competing minima. Spinodal boundaries describe loss of local stability, not coexistence.
An order parameter distinguishes the equilibrium phases and captures the relevant broken symmetry. For an Ising model, it is the spontaneous magnetization: its sign changes under spin reversal, and it is zero in the symmetric phase. Spontaneous magnetization means taking the thermodynamic limit before sending the external field to zero; the finite-volume zero-field mean can vanish even in an ordered phase.
In Landau-Ginzburg theory, a slowly varying real scalar field represents the coarse-grained local order parameter. A useful local free-energy density gives the functional
At zero field, spin-reversal symmetry excludes odd powers. The gradient term penalizes rapid spatial variation. The field is the field conjugate to an order parameter. The mean-field approximation replaces the fluctuating scalar field by the equilibrium minimizer; a uniform minimizer has and obeys .
Consider the complex-linear two-coordinate map . The triangle inequality shows that its operator norm from to is one; equality is attained at . The complex parallelogram identity
shows that its norm from to is . Interpolate these endpoint bounds by the Riesz-Thorin theorem with parameter . Then
Dividing the output norm by proves the scalar estimate:
Apply it pointwise to and integrate its th power. With , this gives
By Minkowski inequality, the expression on the right is at most
Therefore the requested function inequality holds, namely the dual-exponent Clarkson inequality.
Put . Since , choose simple functions of finite-measure support with in . These belong to both endpoint domain spaces. The given simple-function estimate applies also to their differences, so
By completeness of Lp spaces, including , there is with in . Taking limits of the simple-function bound yields .
It remains to identify with the already defined , rather than just define an unrelated extension. Write and with their sum norms. The endpoint estimates imply
Indeed, split , bound the two images in their endpoint norms, and take the infimum over all splits. Part (i) gives , hence in .
The inclusion is also continuous. For distinct finite endpoints this is the same threshold argument as part (i), after ordering the endpoints. If the larger endpoint is infinity, splitting at bounds both the lower endpoint norm of the large-value part and the supremum norm of the small-value part by . If the endpoints coincide, the inclusion is immediate. Thus also in . Uniqueness of limits in the sum norm gives almost everywhere. Consequently the Riesz-Thorin theorem estimate extends as
The interpolated exponent is strictly between and . Apply Hölder's inequality to with conjugate exponents and . Their reciprocals sum to one, so
This gives .
For the other inclusion take any and use the threshold decomposition between Lp spaces
Because ,
Thus , and , proving . Raising the two norm bounds to powers and gives a power of equal to
and a power of equal to one. Hence
For later use, choose when . Each endpoint norm is then at most , so for the sum of Lp spaces norm. If , take .
For , define the forward and backward one-sided interval averaging operators
and set . Each positive-radius average is bounded in absolute value by . Although is an endpoint of its averaging interval, enlarge that interval slightly so that is interior, apply the maximal bound and let the enlargement shrink to zero.
To dominate the limiting identity too without assuming the differentiation result we are proving, use the sublinear operator
The maximal estimate and Chebyshev inequality imply
Thus has weak type (1,1) and dominates every , including .
Smooth compactly supported functions are dense in , obtained by truncating and then smoothing with a compactly supported approximate identity. For each such continuous function both averages tend to its value at every point. Applying part (ii) to the two families, and intersecting their full-measure convergence sets, gives almost everywhere.
The forward difference quotient of the primitive is , and the backward quotient is :
They have the same limit, proving differentiability almost everywhere, with the stronger conclusion
This proves differentiation of an indefinite Lebesgue integral directly from the maximal estimate and density, without assuming as an unproved prerequisite.
Let . It is open: it is the union of all open intervals for which . This also proves measurability of the uncentered maximal function. Triangle inequality inside each average proves sublinearity and absolute homogeneity.
Take a compact subset , and choose finitely many such intervals covering it. The one-dimensional Wiener covering lemma selects pairwise disjoint intervals whose concentric triples cover . For a finite family this follows by choosing a longest remaining interval and discarding all intervals intersecting it: each discarded interval has no greater length and lies in the triple of the chosen one. Thus
Taking the supremum over compact subsets, by inner regularity of Lebesgue measure, gives
This establishes weak type (1,1). It is the one-dimensional uncentered maximal weak-type inequality for the finite measure ; the constant need not be optimal for this question.
A nonnegative sublinear operator satisfies and almost everywhere. It has weak type (1,1) if there is a finite constant , independent of the function and threshold, such that
For a signed or complex output use in the level set. The bound controls the distribution of large values; it does not assert that is integrable.
The covariant Landau gauge is the zero-parameter limit of the covariant gauge family. It imposes the Lorenz gauge condition sharply. The Landau gauge photon propagator is the transverse part of the inverse from solution:
Away from the pole, , matching the divergence-free condition. This is transversality to the four-momentum, rather than a projector onto only the two physical spatial photon polarizations. The Lorenz gauge equation alone still admits residual gauge transformations with ; boundary conditions and the pole prescription complete the propagator choice. The parameter limit specifies the Landau propagator without trying to invert the ungauged Maxwell operator on its longitudinal null space.
The Maxwell quadratic action has a null longitudinal direction: its momentum-space kinetic operator annihilates , reflecting gauge invariance under addition of a gradient to the potential. It cannot be inverted on all four vector components before gauge fixing.
Add the covariant gauge density . Away from the pole, introduce the transverse and longitudinal momentum projectors
With one index raised they obey , , and . The gauge-fixed kinetic operator is . Its inverse times gives the photon propagator
The pole factors are interpreted with the Feynman boundary prescription; the algebraic projector inversion is performed off the pole. The choice is Feynman gauge.
Changing changes only a longitudinal term proportional to . Between conserved currents it vanishes because . In full QED amplitudes, the Ward-Takahashi identity supplies the corresponding cancellations when all diagrams at an order are combined. Thus physical on-shell scattering is gauge-parameter independent, whereas off-shell Green functions and the propagator itself can depend on the gauge choice. For the charged scalar, derivative vertices and contact terms must likewise be included for this cancellation.
A constant phase leaves the photon unchanged and makes every QED interaction vertex invariant: its one and one acquire opposite phases. Thus the propagators and vertex factors in the QED Feynman rules are unchanged. A Green function with insertions of and of acquires the phase . For an invariant vacuum it can be nonzero only when its total inserted charge is zero.
Equivalently, the conserved charge operator commutes with the S-matrix. If initial and final states have charges , then
The scattering charge selection rule from global gauge symmetry is therefore
In diagrams the oriented fermion lines preserve charge at every vertex, and a photon is neutral. Pair creation is allowed because a particle and its antiparticle have opposite charges. Global phase invariance supplies this selection rule; local gauge invariance supplies the stronger momentum-dependent Ward-Takahashi identity.
Choose the phase convention in the question and use the gauge covariant derivative . For a charged Dirac field, the quantum electrodynamics Lagrangian is
The simultaneous local gauge transformation is
Direct differentiation gives : the derivative of the phase cancels the shift of . The mass term and covariant kinetic term are therefore invariant. The electromagnetic field tensor is unchanged because mixed partial derivatives of commute. This proves classical gauge invariance of the coupled field action. A charged scalar similarly uses and a potential depending only on .
The electromagnetic coupling is . The matter equations give a conserved current , with . Constant gives the global symmetry and its conserved charge. When applying Noether's theorem, the sign of the parameterized Noether generator can be reversed without changing this conservation law; the transformation and coupling signs above remain fixed.
Let annual total loss be , with an empty sum equal to zero. The assumptions give a compound Poisson distribution. Conditioning on the count,
The tower property and total-variance formula therefore give
The term is the extra uncertainty in the number of defaults. Replacing it by zero would incorrectly treat that count as deterministic.
For sufficiently many annual events and an adequate normal approximation, the value at risk at confidence level is the compound-Poisson annual loss quantile
The supplied quantiles give , and . In particular a one-sided 95% annual loss quantile uses , not the from the two-sided 95% rate interval. If the expected loss is covered separately, the approximate unexpected-loss reserve is the second term, rather than the full quantile.
Estimate the rate from a suitable historical count/exposure or gap sample, then insert into the loss formula. The interval in part (d) measures rate-estimation uncertainty; for nonnegative loss mean it can be propagated to lower and upper approximate quantiles because the displayed expression is increasing in . Severity parameters also require estimation, so the rate interval alone is not a confidence interval accounting for every source of uncertainty.
Mean and variance do not determine the exact loss quantile. The actual distribution is
For nonnegative losses it has a zero atom at least , so if that mass exceeds its exact value at risk is zero. With a small rate or a strongly skewed severity distribution, a normal approximation can therefore be poor. A loan book with dependent defaults, rating migration or varying exposures also needs more than the constant-intensity independent-severity model. These are model qualifications, not deductions from the credit rating alone.

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!
We have two killer features:
  1. 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-calculus
    Articles 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/derivative
  2. 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.
    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.
  3. https://raw.githubusercontent.com/ourbigbook/ourbigbook-media/master/feature/x/hilbert-space-arrow.png
  4. Infinitely deep tables of contents:
    Figure 6.
    Dynamic article tree with infinitely deep table of contents
    .
    Descendant pages can also show up as toplevel e.g.: ourbigbook.com/cirosantilli/chordate-subclade
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