A new method for obtaining sharp compound Poisson approximation error estimates for sums of locally dependent random variables
Boutsikas, Michael V. ; Vaggelatou, Eutichia
Bernoulli, Tome 16 (2010) no. 1, p. 301-330 / Harvested from Project Euclid
Let X1, X2, …, Xn be a sequence of independent or locally dependent random variables taking values in ℤ+. In this paper, we derive sharp bounds, via a new probabilistic method, for the total variation distance between the distribution of the sum ∑i=1nXi and an appropriate Poisson or compound Poisson distribution. These bounds include a factor which depends on the smoothness of the approximating Poisson or compound Poisson distribution. This “smoothness factor” is of order O(σ−2), according to a heuristic argument, where σ2 denotes the variance of the approximating distribution. In this way, we offer sharp error estimates for a large range of values of the parameters. Finally, specific examples concerning appearances of rare runs in sequences of Bernoulli trials are presented by way of illustration.
Publié le : 2010-05-15
Classification:  compound Poisson approximation,  coupling inequality,  law of small numbers,  locally dependent random variables,  Poisson approximation,  rate of convergence,  total variation distance,  Zolotarev’s ideal metric of order 2
@article{1274821072,
     author = {Boutsikas, Michael V. and Vaggelatou, Eutichia},
     title = {A new method for obtaining sharp compound Poisson approximation error estimates for sums of locally dependent random variables},
     journal = {Bernoulli},
     volume = {16},
     number = {1},
     year = {2010},
     pages = { 301-330},
     language = {en},
     url = {http://dml.mathdoc.fr/item/1274821072}
}
Boutsikas, Michael V.; Vaggelatou, Eutichia. A new method for obtaining sharp compound Poisson approximation error estimates for sums of locally dependent random variables. Bernoulli, Tome 16 (2010) no. 1, pp.  301-330. http://gdmltest.u-ga.fr/item/1274821072/