×

On the almost sure convergence of floating-point mantissas and Benford’s law. (English) Zbl 0645.60038

Let \(Y_ 1\), \(Y_ 2\),... be a sequence of random variables and let \(M_ n\) be the floating-point mantissa of \(Y_ n\). Further let \(1_{[1,x)}(\cdot)\) denote the indicator function of the interval [1,x). If \(Y_ n/n\to Z\) a.s., where \(Z\neq 0\) is a further random variable, then the sequence \(1_{[1,x)}(M_ n)\) converges a.s. to log x in the sense of \({\mathcal H}\infty\)-means and logarithmic means, respectively. The speed of convergence in this relations is estimated. As a conclusion, a further argument for Benford’s law is provided.

MSC:

60F15 Strong limit theorems
Full Text: DOI

References:

[1] Barlow, Computing 34 pp 325– (1985)
[2] Benford, Proc. Amer. Phil. Soc. 78 pp 551– (1938)
[3] Flehinger, Amer. Math. Monthly 73 pp 1056– (1966)
[4] Goodman, Computing 15 pp 263– (1975)
[5] Holewijn, Z. Wahrsch. Verw. Gebiete 14 pp 89– (1969)
[6] Raimi, Amer. Math. Monthly 83 pp 521– (1976)
[7] Robbins, Proc. Amer. Math. Soc. 4 pp 786– (1953)
[8] Schatte, Zeitschr. f. Angew. Math. und Mech. 53 pp 553– (1973)
[9] Schatte, Wiss. Zeitschr. d. Univ. Rostock 23 pp 783– (1974)
[10] Schatte, Monatsh. Math. 103 pp 233– (1987)
[11] Schatte, EIK J. Inf. Process. Cybern. EIK 23 pp 353– (1987)
[12] Schatte, Probab. Theory Related Fields (1988)
[13] und , Theorie der Limitierungsverfahren, Springer Berlin 1970 · doi:10.1007/978-3-642-88470-2
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.