Almost everywhere exponential convergence of the modified Jacobi—Perron algorithm (Q3140464)
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 437583
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | Almost everywhere exponential convergence of the modified Jacobi—Perron algorithm |
scientific article; zbMATH DE number 437583 |
Statements
Almost everywhere exponential convergence of the modified Jacobi—Perron algorithm (English)
0 references
23 September 1996
0 references
simultaneous approximation
0 references
ergodic transformation
0 references
Jacobi-Perron algorithm
0 references
0.7219044
0 references
0.72114706
0 references
0.71915466
0 references
0.7112926
0 references
0.70074826
0 references
0 references
0.6941804
0 references
The modified Jacobi-Perron algorithm is associated to a transformation \(T\) on the unit square \(X\), defined by \(T(\alpha, \beta)= (\beta/\alpha, \{1/ \alpha\}))\) if \(\beta< \alpha\) and \(T(\alpha, \beta)= (\{1/ \beta\}, \alpha/\beta)\) if \(\alpha< \beta\). Here \(\{\cdot\}\) stands for the fractional part and \(T\) is not defined for \(\alpha \beta= 0\) or \(\alpha= \beta\). If the algorithm is performed \(n\)-times, we get integers \(p_n\), \(p_n'\), \(p'_n\), \(r_n\), \(r_n'\), \(r'_n\), \(q_n\), \(q_n'\), \(q'_n\) such that \(\alpha= {p_n+ p_n' \alpha_n+ p'_n \beta_n\over q_n+ q_n' \alpha_n+ q'_n \beta_n}\) and \(\beta= {r_n+ r_n' \alpha_n+ r'_n \beta_n\over q_n+ q_n' \alpha_n+ q'_n \beta_n}\) with \((\alpha_n, \beta_n)= T^n(\alpha, \beta)\). In this paper, the authors prove that there exists a constant \(\delta> 0\) such that for Lebesgue almost every \((\alpha, \beta)\) in \(X\) that exists \(n_0\) (depending on \((\alpha, \beta)\)) such that for all \(n\geq n_0\). \[ \Biggl|\alpha- {p_n\over q_n} \Biggr|< {1\over q^{1+ \delta}_n},\quad \Biggl|\beta- {r_n\over q_n}\Biggr|< {1\over q^{1+ \delta}_n}. \] The proof uses the explicit formula of the absolutely continuous \(T\)-invariant ergodic probability measure, given by \textit{F. Schweiger} [``The metrical theory of the Jacobi-Perron algorithm'', Lect. Notes Math. 334 (1973; Zbl 0287.10041)]. By applying the subadditive ergodic theorem, the inequality \(\delta> 0\) is reduced to an extensive calculation performed manually.
0 references