Sample size calculation for comparing two normal random samples
using equilibrium priors
N. T. Longford
Abstract
A method of sample size calculation for deciding
whether the difference of the expectations of two normal random samples
is positive is introduced.
Prior information is used, in the form of a set of plausible prior
distributions and plausible loss functions for the two kinds
of incorrect decision.
The sample size is set so that the prior-related probability
of concluding the analysis with an impasse is smaller than a given threshold
or by minimising the total of the cost of the planned experiment
and the expected losses due to incorrect decision
and inconclusive outcome of the analysis.
Submitted.
April 2011.