Bias Correction of Sample Proportion for Length or Size Biased Data

DOI: https://doi.org/10.59321/BAUETJ.V4I1.1

AUTHOR(S)
Noora N Saleh1, Keya Rani Das2 and A.H.M. Rahmatullah Imon3*

ABSTRACT
In many areas of statistics it is not easy to collect samples which are entirely random in nature. When we collect data from nature by encounter and/or grab sampling we often observe that the collected data are length and/or size biased. The main limitation of this type of data is that they are biased. If we can estimate the bias factor then we can do bias correction and after that the resulting estimates are considered as almost equally reliable to their unbiased counterparts. Bias correction of mean for length and/or size biased data is available in the literature. This is of course a great development but often we are more interested to estimate the proportion rather than the mean. For example, we may be more interested to know the proportion of healthy plants in a garden or the proportion of products which comply the standard. In this paper we develop a bias corrected estimate of proportion for length and/or size biased data. We study few of their mathematical properties and its application in the real world situation.

Download Full Article