Asymptotic behaviour of the first moment of the number of steps in the by-excess and by-deficiency Euclidean algorithms (Q2884667)

From MaRDI portal

!

This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:

scientific article; zbMATH DE number 6039351
Language Label Description Also known as
default for all languages
No label defined
    English
    Asymptotic behaviour of the first moment of the number of steps in the by-excess and by-deficiency Euclidean algorithms
    scientific article; zbMATH DE number 6039351

      Statements

      Asymptotic behaviour of the first moment of the number of steps in the by-excess and by-deficiency Euclidean algorithms (English)
      0 references
      30 May 2012
      0 references
      Euclidean algorithms
      0 references
      continued fractions
      0 references
      fractional parts
      0 references
      prime number theorem
      0 references
      0 references
      For \(a/b\in(0,1]\) denote by \(s(a/b)\) the length of the standard continued fraction expansion NEWLINE\[NEWLINE\frac{a}{b}=\frac{1}{a_1+{\atop\ddots\,\displaystyle{+\frac{1}{a_s}}}},NEWLINE\]NEWLINE and let \(l(a/b)\) be a length of reduced regular continued fraction NEWLINE\[NEWLINE\frac{a}{b}=1-\frac{1}{b_1-{\atop\ddots\,\displaystyle{-\frac{1}{b_l}}}}\qquad(b_1, \ldots, b_l\geq 2).NEWLINE\]NEWLINE It was known (see [\textit{A. V. Ustinov}, Izv. Math. 72, No. 5, 1023--1059 (2008); translation from Izv. Ross. Akad. Nauk, Ser. Mat. 72, No. 5, 189--224 (2008; Zbl 1211.11009)]; [\textit{E. N. Zhabitskaya}, Math. Notes 89, No. 3, 450--454 (2011); translation from Mat. Zametki 89, No. 3, 472--472 (2011; Zbl 1306.11010)]) that average lengths NEWLINE\[NEWLINEE^+(R)=\frac{2}{[R]([R]+1)}\sum\limits_{b\leq R}\sum\limits_{a\leq b}s\left(\frac{a}{b}\right),\qquad E^-(R)=\frac{2}{[R]([R]+1)}\sum\limits_{b\leq R}\sum\limits_{a\leq b}l\left(\frac{a}{b}\right)NEWLINE\]NEWLINE satisfy asymptotic formulae NEWLINE\[NEWLINEE^+(R)=c_1^+\log R+c_0^++O(\omega^+(R)),\qquad E^-(R)=c_2^+\log^2 R+c_1^+\log R+c_0^++O(\omega^-(R))NEWLINE\]NEWLINE with some explicit constants \(c_i^\pm\) and the error terms NEWLINE\[NEWLINE\omega^\pm(R)=O(R^{-1}\log^5R).NEWLINE\]NEWLINE The author proves sharper estimates NEWLINE\[NEWLINE\omega^+(R)=O(R^{-1}\log^3R),\tag{1}NEWLINE\]NEWLINE NEWLINE\[NEWLINE\omega^-(R)=O(R^{-1}\log^3R). \tag{2}NEWLINE\]NEWLINE It is interesting that the proof of the estimate (2) uses some ideas from Selberg's elementary proof of the prime number theorem.NEWLINENEWLINEEstimates (1) and (2) are nearly optimal, although \(\Omega\)-theorems for \(\omega^\pm(R)\) are not known. According to numerical experiments performed by the author, the error terms \(\omega^\pm(R)\) should satisfy the estimates NEWLINE\[NEWLINE\omega^+(R)=O(R^{-1}\log R),\qquad \omega^-(R)=O(R^{-1}\log^2 R).NEWLINE\]
      0 references

      Identifiers