Year: 2015
Type: preprint
Abstract: I introduce a completely new (non-parametric) approach to determine (satislcing) sample sizes, when dynamic sampling and stopping rules are not feasible, and when no assumption can be made on the underlying (and unknown) distribution(s) that is (are) believed to generate observed experimental data. The method proposed, that relies on the construction of a 'binding function', is shown to be a general solution for a class of problems associated with decision functions that frequently interest (lab and leld) experimental researchers in the areas of economics and psychology.
Author: Subrato Banerjee
Language: English
Subfield: Statistics and Probability
Field: Mathematics
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W2917224715
