On continued fraction randomness and normality (Q2672257): Difference between revisions

The present paper deals with the concept of an individual random continued fraction. In this research, the main problem is ``whether randomness of a real is preserved when translating from one representation to another''. Special attention is given to a brief survey in this topic, to useful notions, as well as to useful estimates for continued fractions and the Gauss measure, to martingales on continued fraction expansions, and to some open problems. One can note authors' abstract: ``Recently, \textit{A.-M. Scheerer} [``On the continued fraction expansion of absolutely normal numbers'', Preprint, \url{arXiv:1701.07979}] and \textit{J. Vandehey} [Bull. Aust. Math. Soc. 94, No. 2, 217--223 (2016; Zbl 1391.11094)] showed that normality for continued fraction expansions and base-\(b\) expansions are incomparable notions. This shows that at some level, randomness for continued fractions and binary expansion are different statistical concepts. In contrast, we show that the continued fraction expansion of a real is computably random if and only if its binary expansion is computably random. To quantify the degree to which a continued fraction fails to be effectively random, we define the effective Hausdorff dimension of individual continued fractions, explicitly constructing continued fractions with dimension \(0\) and \(1\).''