LOWER BOUNDS FOR DISCRETE APPROXIMATIONS TO SUMS OF
\(m\)-DEPENDENT RANDOM VARIABLES
Vydas Cekanavicius
Palaniappan Vellaisamy
Abstract:
Lower bounds for the second order Poisson, compound Poisson,
negative binomial and binomial approximations to the sum of
1-dependent random variables are obtained for the Kolmogorov and local metrics. The results are then applied to sums of independent
indicators and runs statistics.
2010 AMS Mathematics Subject Classification: Primary 60F05; Secondary 60G50.
Keywords and phrases: Poisson
approximation, signed compound Poisson measure, binomial
and negative binomial approximations, one-dependent
variables, local norm, Kolmogorov distance.