Formal Analysis

Approximation orders of formal Laurent series by β-expansions

In this paper, the constrained dispersionless BKP hierarchy are given by restricting the formal Laurent series [Formula: see text] of the dispersionless BKP hierarchy to [Formula: see text]. Its additional symmetries are given by introducing vital formal Laurent series [Formula: see text]. The additional flows form the nonabelian [Formula: see text]-grading Lie algebra. Furthermore, the additional flows acting on [Formula: see text] and [Formula: see text] are presented.

A higher-dimensional Kurzweil theorem for formal Laurent series over finite fields

Consider a polynomial ring R with the ‐grading where the degree of each variable is a standard basis vector. In other words, R is the homogeneous coordinate ring of a product of n projective spaces. In this setting, we characterize the formal Laurent series which arise as Hilbert series of finitely generated R‐modules. We also provide necessary conditions for a formal Laurent series to be the Hilbert series of a finitely generated module with a given depth. In the bigraded case (corresponding to the product of two projective spaces), we completely classify the Hilbert series of finitely generated modules of positive depth.

Helicon wave (whistler waves bounded transversely by a magnetic field surface) propagation is investigated for a cylindrically symmetric curved (flaring) boundary using a finite-element method where the propagation region is divided into a sequence of truncated cones. In each conical segment, a local spherical coordinate system is used with the origin at the apex of the cone. A vector wave equation for the fields is formulated for a cold plasma and reduced, in spherical coordinates, to a pair of coupled partial differential equations (PDEs) for two scalar functions. The PDEs are separable for the azimuthal eigenvalue m = 0. The -dependence is a Legendre function, with non-integer eigenvalues determined by the cone angle. The dependence on the radial coordinate, x, satisfies a fourth-order ordinary differential equation (ODE). A straightforward numerical integration of the equation from to fails for large values of , because of the existence of an exponentially growing solution to the equation. Consequently, a different approach is needed. Four independent solutions, valid for , in the form of power series in x (PS) are obtained which are each asymptotically proportional to as . For x large, four asymptotic expansions (AE) are obtained in the form of series in 1/x times or . The problem then is to find four linear combinations of the four PSs, each of which approaches one of the AEs in the limit by first deriving a double-integral representation of each PS, valid for all x, and then using its limit to match each AE to a linear combination of the PSs. For special values of or 4N, N an integer) closed-form solutions (polynomials times exponentials) result that are exact representations of the solution for all x. Solutions are computed and illustrated for an outgoing wave using the closed-form solution for special values of , and as a function of and x using the PS and AE, which have a large range of overlap. The propagating solutions of the ODE are also obtained using a WKB method and are used to calculate the propagation in a slowly diverging parabola of revolution.

We investigate the translation equation (T) , and the cocycle equations (Co1) , (Co2) in [ x ], the ring of formal power series over . Here we restrict ourselves to iteration groups of type I, i. e. to solutions of (T) of the form , where c 1 ? 1 is a generalized exponential function. It is easy to prove that the coefficient functions c n (s), a n (s), s n (s) of are polynomials, c 1(s), polynomials or rational functions in one more generalized exponential function and, in a special case, polynomials in an additive function. For example, we obtain . For this we do not need any detailed information about the polynomials P n . Under some conditions on the exponential and additive functions it is possible to replace the exponential and additive functions by independent indeterminates. In this way we obtain formal versions of the translation equation and the cocycle equations in rings of the form [ x ], [ x ], [ x ], [ x ], and [ x ]. We solve these equations in a completely algebraic way, by deriving formal differential equations or Aczel�Jabotinsky type equations for these problems. It is possible to get the structure of the coefficients in great detail which are now polynomials or rational functions. We prove the universal character (depending on certain parameters) of these polynomials or rational functions. And we deduce the canonical form S -1(yS(x)) for iteration groups of type I. This approach seems to be more clear and more general than the original one. Some simple substitutions allow us to solve these problems in rings of the form ( [ u ]) [ x ], i. e. where the coefficient functions are formal series.

A positive cosmological constant simplifies the asymptotics of forever expanding cosmological solutions of the Einstein equations. In this paper a general mathematical analysis on the level of formal power series is carried out for vacuum spacetimes of any dimension and perfect fluid spacetimes with linear equation of state in spacetime dimension four. For equations of state stiffer than radiation evidence for development of large gradients, analogous to spikes in Gowdy spacetimes, is found. It is shown that any vacuum solution satisfying minimal asymptotic conditions has a full asymptotic expansion given by the formal series. In four spacetime dimensions, and for spatially homogeneous spacetimes of any dimension, these minimal conditions can be derived for appropriate initial data. Using Fuchsian methods the existence of vacuum spacetimes with the given formal asymptotics depending on the maximal number of free functions is shown without symmetry assumptions.

defined over the field of rational numbers Q is normal. They do this by noting that, over a field of characteristic 3, F (X) equals 1/ √ 1 + 1/X. The latter has a periodic continued fraction and, as a consequence, they are able to explicitly verify that F (X) is normal. They then use an “automatic” proof to show that the Laurent series of F (X), defined over Q, is normal. Here we shall give a different proof of the latter result (Theorem 4, below) and show, more generally, that if a formal Laurent series Γ (X) defined over Q is defined and normal (mod p) for any prime p, then it is also normal over Q. We shall also consider the continued fraction of the more general surd 1/ √ 1 + u/X + v/X2 over any field not of characteristic 2. Put δ = u/4−v and note that √ 1 + u/X + v/X2 = 1 + u/(2X) when δ = 0. Hence, when δ = 0, the continued fraction expansion is trivial and is not normal. Define

Formal power series and regular operations on fuzzy languages

Pseudodifferential operators of several variables are formal Laurent series in the formal inverses of ∂ 1 , …, ∂ n with ∂ i=d/dx i for 1≤i ≤ n. As in the single variable case, Lax equations can be constructed using such pseudodifferential operators, whose solutions can be provided by Baker functions. We extend the usual notion of tau functions to the case of pseudodifferential operators of several variables so that each Baker function can be expressed in terms of the corresponding tau function.

We investigate the nature of divergences in quantum field theory, showing that they are organized in the structure of a certain “motivic Galois group” U*, which is uniquely determined and universal with respect to the set of physical theories. The renormalization group can be identified canonically with a one-parameter subgroup of U*. The group U* arises through a Riemann-Hilbert correspondence. Its representations classify equisingular flat vector bundles, where the equisingularity condition is a geometric formulation of the fact that in quantum field theory the counterterms are independent of the choice of a unit of mass. As an algebraic group scheme, U* is a semidirect product by the multiplicative group G_m of a prounipotent group scheme whose Lie algebra is freely generated by one generator in each positive integer degree. There is a universal singular frame in which all divergences disappear. When computed as iterated integrals, its coefficients are certain rational numbers that appear in the local index formula of Connes-Moscovici. When working with formal Laurent series over ℚ, the data of equisingular flat vector bundles define a Tannakian category whose properties are reminiscent of a category of mixed Tate motives.

We prove an explicit formula for infinitely many convergents of Hurwitzian continued fractions that repeat several copies of the same constant and elements of one arithmetic progression, in a quasi-periodic fashion. The proof involves combinatorics and formal Laurent series. Using very little analysis we can express their limits in terms of (modified) Bessel functions and Fibonacci polynomials. The limit formula is a generalization of Lehmer's theorem that implies the continuous fraction expansions of $e$ and $\tan(1)$, and it can also be derived from Lehmer's work using Fibonacci polynomial identities. We completely characterize those implementations of our limit formula for which the parameter of each Bessel function is the half of an odd integer, allowing them to be replaced with elementary functions.

Let p be a prime. This paper deals with solutions of functional equations in either formal Laurent series or in analytic functions. Examples connected to special modular functions are considered.

Substituting non-unit formal power series into formal power series is a well-understood concept. This article describes a sense in which one can define substitution of unit formal power series, implemented in Maple’s MultivariatePowerSeries package.

Let S n be the symmetric group, T d be the subspace of C[x 1 , x 2 ,…,x n ] which contains all of the homogeneous polynomials of degree d. The representation of S n determined by T d is denoted by λ d , and the irreducible representation characterized by partition a is denoted by π(a). We denote the multiplicity of π(a) entries into λ d by N a d . As an extension of [4], this note simplifies the recursion formula of the formal power series (?)N a d t d , and obtains the explicity expressions of generating function...