A Sampling Algorithm for Generating Representative Random Samples given Small Sample Size

A novel algorithm for sampling from discrete probability distributions

A novel algorithm is developed for sampling from discrete probability distributions using the probability proportional to size sampling method, which is a special case of Quota sampling method.

The goal is to devise an efficient sampling algorithm that can be used in stochastic optimization problems -- when there is a need to minimize the sample size.

The basic idea of the proposed algorithm is as follows: For a fixed number of random samples N and either ⌊Nx_i ⌋ or ⌈Nx_i ⌉ total observations for the x_i value; the distribution is determined over [⌊Nx_1 ⌋ ,⌈Nx_1 ⌉ ] ,… ,[⌊Nx_k ⌋ ,⌈Nx_k ⌉ ] that best fits the original distribution. More specifically, all possible distributions over[⌊Nx_1 ⌋ ,⌈Nx_1 ⌉ ] ,… ,[⌊Nx_k ⌋ ,⌈Nx_k ⌉ ]are generated; then the distributions that don’t satisfy the total sample size, N, are excluded. After that the best fit distribution is selected

For further details, kindly read the following paper:
http://www.igi-global.com/article/novel-quota-sampling-algorithm-generating/76350