The cut-off point based on underlying distribution and cost function

Abstract

Cut-off sampling has been widely used for business survey which has the right-skewed population with a long tail. Several methods are suggested to obtain the optimal cut-off point. The LH algorithm suggested by Lavallee and Hidiroglou [6] is commonly used to get the optimum boundaries by minimizing the total sample size with a given precision. In this paper, we suggest a new cut-off point determination method which minimizes a cost function. And that leads to reducing the size of take-all stratum. Also we investigate an optimal cut-off point using a typical parametric estimation method under the assumptions of underlying distributions. Small Monte-Carlo simulation studies are performed in order to compare the new cut-off point method to the LH algorithm. The Korea Transportation Origin – Destination data are used for real data analysis.

Keywords:

• truncated log-normal distribution
• truncated gamma distribution
• take-all stratum
• take-some stratum
• LH algorithm

Disclosure statement

No potential conflict of interest was reported by the authors.

Funding

This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education science and Technology [2012R1A1A2003919].