Showing 1–6 of 6 results for author: Athreya, K B

Extrema of Luroth Digits and a zeta function limit relation

Authors: Jayadev S. Athreya, Krishna B. Athreya

Abstract: We describe how certain properties of the extrema of the digits of Luroth expansions lead to a probabilistic proof of a limiting relation involving the Riemann zeta function and the Bernoulli triangles. We also discuss trimmed sums of Luroth digits. Our goal is to show how direct computations in this case lead to explicit formulas and some interesting discussions of special functions. We describe how certain properties of the extrema of the digits of Luroth expansions lead to a probabilistic proof of a limiting relation involving the Riemann zeta function and the Bernoulli triangles. We also discuss trimmed sums of Luroth digits. Our goal is to show how direct computations in this case lead to explicit formulas and some interesting discussions of special functions. △ Less

Submitted 3 October, 2021; originally announced October 2021.

Comments: to appear in Integers (http://math.colgate.edu/~integers/)

MSC Class: 60G70; 11M06; 11B68

AR(1) sequence with random coefficients: Regenerative properties and its application

Authors: Krishna B. Athreya, Koushik Saha, Radhendushka Srivastava

Abstract: Let $\{X_n\}_{n\ge0}$ be a sequence of real valued random variables such that $X_n=ρ_n X_{n-1}+ε_n,~n=1,2,\ldots$, where $\{(ρ_n,ε_n)\}_{n\ge1}$ are i.i.d. and independent of initial value (possibly random) $X_0$. In this paper it is shown that, under some natural conditions on the distribution of $(ρ_1,ε_1)$, the sequence $\{X_n\}_{n\ge0}$ is regenerative in the sense that it could be broken up i… ▽ More Let $\{X_n\}_{n\ge0}$ be a sequence of real valued random variables such that $X_n=ρ_n X_{n-1}+ε_n,~n=1,2,\ldots$, where $\{(ρ_n,ε_n)\}_{n\ge1}$ are i.i.d. and independent of initial value (possibly random) $X_0$. In this paper it is shown that, under some natural conditions on the distribution of $(ρ_1,ε_1)$, the sequence $\{X_n\}_{n\ge0}$ is regenerative in the sense that it could be broken up into i.i.d. components. Further, when $ρ_1$ and $ε_1$ are independent, we construct a non-parametric strongly consistent estimator of the characteristic functions of $ρ_1$ and $ε_1$. △ Less

Submitted 12 September, 2017; originally announced September 2017.

MSC Class: 60G20

Local times in a Brownian excursion

Authors: Krishna B. Athreya, Raoul Normand, Vivekananda Roy, Sheng-Jhih Wu

Abstract: Let $\{B(t), t \geq 0\}$ be a standard Brownian motion in $\mathbb{R}$. Let $T$ be the first return time to 0 after hitting 1, and $\{L(T,x), x \in \mathbb{R}\}$ be the local time process at time $T$ and level $x$. The distribution of $L(T,x)$ for each $x \in \mathbb{R}$ is determined. This is applied to the estimation of a $L^1$ integral on $\mathbb{R}$. Let $\{B(t), t \geq 0\}$ be a standard Brownian motion in $\mathbb{R}$. Let $T$ be the first return time to 0 after hitting 1, and $\{L(T,x), x \in \mathbb{R}\}$ be the local time process at time $T$ and level $x$. The distribution of $L(T,x)$ for each $x \in \mathbb{R}$ is determined. This is applied to the estimation of a $L^1$ integral on $\mathbb{R}$. △ Less

Submitted 17 October, 2014; originally announced October 2014.

Comments: 8 pages

MSC Class: Primary: 60J65; Secondary: 60F05

T. E. Harris and branching processes

Authors: K. B. Athreya, P. E. Ney

Abstract: T. E. Harris was a pioneer par excellence in many fields of probability theory. In this paper, we give a brief survey of the many fundamental contributions of Harris to the theory of branching processes, starting with his doctoral work at Princeton in the late forties and culminating in his fundamental book "The Theory of Branching Processes," published in 1963. T. E. Harris was a pioneer par excellence in many fields of probability theory. In this paper, we give a brief survey of the many fundamental contributions of Harris to the theory of branching processes, starting with his doctoral work at Princeton in the late forties and culminating in his fundamental book "The Theory of Branching Processes," published in 1963. △ Less

Submitted 10 March, 2011; originally announced March 2011.

Comments: Published in at http://dx.doi.org/10.1214/10-AOP599 the Annals of Probability (http://www.imstat.org/aop/) by the Institute of Mathematical Statistics (http://www.imstat.org)

Report number: IMS-AOP-AOP599

Journal ref: Annals of Probability 2011, Vol. 39, No. 2, 429-434

Continuity of Translation Operators

Authors: Krishna B. Athreya, Justin R. Peters

Abstract: For a Radon measure $μ$ on $\bbR,$ we show that $L^{\infty}(μ)$ is invariant under the group of translation operators $T_t(f)(x) = {$f(x-t)$}\ (t \in \bbR)$ if and only if $μ$ is equivalent to Lebesgue measure $m$. We also give necessary and sufficient conditions for $L^p(μ),\1 \leq p < \infty,$ to be invariant under the group $\{T_t\}$ in terms of the Radon-Nikodym derivative w.r.t. $m$. For a Radon measure $μ$ on $\bbR,$ we show that $L^{\infty}(μ)$ is invariant under the group of translation operators $T_t(f)(x) = {$f(x-t)$}\ (t \in \bbR)$ if and only if $μ$ is equivalent to Lebesgue measure $m$. We also give necessary and sufficient conditions for $L^p(μ),\1 \leq p < \infty,$ to be invariant under the group $\{T_t\}$ in terms of the Radon-Nikodym derivative w.r.t. $m$. △ Less

Submitted 31 October, 2010; originally announced November 2010.

Comments: to appear, PAMS

MSC Class: primary: 26A42. secondary: 28A25; 22F10

Growth of preferential attachment random graphs via continuous-time branching processes

Authors: K. B. Athreya, A. P. Ghosh, S. Sethuraman

Abstract: A version of ``preferential attachment'' random graphs, corresponding to linear ``weights'' with random ``edge additions,'' which generalizes some previously considered models, is studied. This graph model is embedded in a continuous-time branching scheme and, using the branching process apparatus, several results on the graph model asymptotics are obtained, some extending previous results, such… ▽ More A version of ``preferential attachment'' random graphs, corresponding to linear ``weights'' with random ``edge additions,'' which generalizes some previously considered models, is studied. This graph model is embedded in a continuous-time branching scheme and, using the branching process apparatus, several results on the graph model asymptotics are obtained, some extending previous results, such as growth rates for a typical degree and the maximal degree, behavior of the vertex where the maximal degree is attained, and a law of large numbers for the empirical distribution of degrees which shows certain ``scale-free'' or ``power-law'' behaviors. △ Less

Submitted 23 January, 2007; originally announced January 2007.

Comments: 20 pages

MSC Class: 05C80; 60J85