English

Generalised regular variation of arbitrary order

Classical Analysis and ODEs 2009-01-13 v1

Abstract

Let ff be a measurable, real function defined in a neighbourhood of infinity. The function ff is said to be of generalised regular variation if there exist functions h≢0h \not\equiv 0 and g>0g > 0 such that f(xt)f(t)=h(x)g(t)+o(g(t))f(xt) - f(t) = h(x) g(t) + o(g(t)) as tt \to \infty for all x(0,)x \in (0, \infty). Zooming in on the remainder term o(g(t))o(g(t)) leads eventually to a relation of the form f(xt)f(t)=h1(x)g1(t)+...+hn(x)gn(t)+o(gn(t))f(xt) - f(t) = h_1(x) g_1(t) + ... + h_n(x) g_n(t) + o(g_n(t)), each gig_i being of smaller order than its predecessor gi1g_{i-1}. The function ff is said to be generalised regularly varying of order nn with rate vector \g=(g1,>...,gn)\g = (g_1, >..., g_n)'. Under general assumptions, \g\g itself must be regularly varying in the sense that \g(xt)=x\B\g(t)+o(gn(t))\g(xt) = x^{\B} \g(t) + o(g_n(t)) for some upper triangular matrix \B\RRn×n\B \in \RR^{n \times n}, and the vector of limit functions \h=(h1,>...,hn)\h = (h_1, >..., h_n) is of the form \h(x)=¸1xu\Bu1\du\h(x) = \c \int_1^x u^\B u^{-1} \du for some row vector ¸\RR1×n\c \in \RR^{1 \times n}. The usual results in the theory of regular variation such as uniform convergence and Potter bounds continue to hold. An interesting special case arises when all the rate functions gig_i are slowly varying, yielding Π\Pi-variation of order nn, the canonical case being that \B\B is equivalent to a single Jordan block with zero diagonal. The theory is applied to a long list of special functions.

Keywords

Cite

@article{arxiv.0901.1468,
  title  = {Generalised regular variation of arbitrary order},
  author = {Edward Omey and Johan Segers},
  journal= {arXiv preprint arXiv:0901.1468},
  year   = {2009}
}

Comments

submitted to the London Mathematical Society

R2 v1 2026-06-21T11:59:35.218Z