TOOLS FOR THE SYMBOLIC COMPUTATION OF ASYMPTOTIC EXPANSIONS

This paper describes a collection of procedures for the systematic com putation of asymptotic expansions that are common in statistical theor y and practice: expansions of functions of sums of independent and ide ntically distributed random variables. The procedures permit the expan sion of maximum likelihood estimates, the associated deviance or drop in likelihood and more general functions of random variables with dist ributions involving one or more parameters. The procedures are illustr ated with examples involving general and specific laws.