Showing 1–3 of 3 results for author: Erhardsson, T

Normal approximation for mixtures of normal distributions and the evolution of phenotypic traits

Authors: Krzysztof Bartoszek, Torkel Erhardsson

Abstract: An explicit bound is given for the Kolmogorov distance between a mixture of normal distributions and a normal distribution with properly chosen parameter values. A random variable X has a mixture of normal distributions if its conditional distribution given some sigma-algebra is normal. The bound depends only on the first two moments of the first two conditional moments of X given this sigma-algeb… ▽ More An explicit bound is given for the Kolmogorov distance between a mixture of normal distributions and a normal distribution with properly chosen parameter values. A random variable X has a mixture of normal distributions if its conditional distribution given some sigma-algebra is normal. The bound depends only on the first two moments of the first two conditional moments of X given this sigma-algebra. As an application, the Yule-Ornstein-Uhlenbeck model, used in the field of phylogenetic comparative methods, is considered. A bound is derived for the Kolmogorov distance between the distribution of the average value of a phenotypic trait over n related species and a normal distribution. The bound goes to 0 as n goes to infinity, extending earlier limit theorems by Bartoszek and Sagitov. △ Less

Submitted 6 August, 2020; v1 submitted 7 December, 2019; originally announced December 2019.

MSC Class: 62E17; 60F05; 92D15

Conditions for convergence of random coefficient AR(1) processes and perpetuities in higher dimensions

Authors: Torkel Erhardsson

Abstract: A $d$-dimensional RCA(1) process is a generalization of the $d$-dimensional AR(1) process, such that the coefficients $\{M_t;t=1,2,\ldots\}$ are i.i.d. random matrices. In the case $d=1$, under a nondegeneracy condition, Goldie and Maller gave necessary and sufficient conditions for the convergence in distribution of an RCA(1) process, and for the almost sure convergence of a closely related sum o… ▽ More A $d$-dimensional RCA(1) process is a generalization of the $d$-dimensional AR(1) process, such that the coefficients $\{M_t;t=1,2,\ldots\}$ are i.i.d. random matrices. In the case $d=1$, under a nondegeneracy condition, Goldie and Maller gave necessary and sufficient conditions for the convergence in distribution of an RCA(1) process, and for the almost sure convergence of a closely related sum of random variables called a perpetuity. We here prove that under the condition $\Vert {\prod_{t=1}^nM_t}\Vert \stackrel{\mathrm{a.s.}}{\longrightarrow}0$ as $n\to\infty$, most of the results of Goldie and Maller can be extended to the case $d>1$. If this condition does not hold, some of their results cannot be extended. △ Less

Submitted 13 March, 2014; originally announced March 2014.

Comments: Published in at http://dx.doi.org/10.3150/13-BEJ513 the Bernoulli (http://isi.cbs.nl/bernoulli/) by the International Statistical Institute/Bernoulli Society (http://isi.cbs.nl/BS/bshome.htm)

Report number: IMS-BEJ-BEJ513

Journal ref: Bernoulli 2014, Vol. 20, No. 2, 990-1005

Strong memoryless times and rare events in Markov renewal point processes

Authors: Torkel Erhardsson

Abstract: Let W be the number of points in (0,t] of a stationary finite-state Markov renewal point process. We derive a bound for the total variation distance between the distribution of W and a compound Poisson distribution. For any nonnegative random variable ζ, we construct a ``strong memoryless time'' \hat ζsuch that ζ-t is exponentially distributed conditional on {\hat ζ\leq t, ζ>t}, for each t. This… ▽ More Let W be the number of points in (0,t] of a stationary finite-state Markov renewal point process. We derive a bound for the total variation distance between the distribution of W and a compound Poisson distribution. For any nonnegative random variable ζ, we construct a ``strong memoryless time'' \hat ζsuch that ζ-t is exponentially distributed conditional on {\hat ζ\leq t, ζ>t}, for each t. This is used to embed the Markov renewal point process into another such process whose state space contains a frequently observed state which represents loss of memory in the original process. We then write W as the accumulated reward of an embedded renewal reward process, and use a compound Poisson approximation error bound for this quantity by Erhardsson. For a renewal process, the bound depends in a simple way on the first two moments of the interrenewal time distribution, and on two constants obtained from the Radon-Nikodym derivative of the interrenewal time distribution with respect to an exponential distribution. For a Poisson process, the bound is 0. △ Less

Submitted 6 October, 2004; originally announced October 2004.

Comments: Published by the Institute of Mathematical Statistics (http://www.imstat.org) in the Annals of Probability (http://www.imstat.org/aop/) at http://dx.doi.org/10.1214/009117904000000054

Report number: IMS-AOP-AOP286 MSC Class: 60K15 (Primary) 60E15 (Secondary)

Journal ref: Annals of Probability 2004, Vol. 32, No. 3B, 2446-2462