Estimations of population parameters in the presence of censored and truncated samples : theory and methods
Year: 2010
Type: article
Abstract: Censored and truncated data are becoming more common in environmental, toxicological, and biomedical studies. This paper is concerned with the problem of estimating the mean and standard deviation of normal populations from truncated, and from both censored and truncated samples. Maximum likelihood estimating equations are derived which can easily be solved by simple iterative processes. Easily computed estimates of the population mean and standard deviation are obtained from (1) truncated normally distributed samples, (2) left truncated and left censored normally distributed samples, and (3) left truncated and right censored normally distributed samples. A new algorithm for obtaining the maximum likelihood estimates of µ and σ from singly-truncated samples is provided which does not require auxiliary tables. The algorithm utilizes the R language. Worked examples are given illustrating the use of these estimators utilizing computer programs whose codes are given in the Appendix.
Author: Abou El-Makarim Aboueissa
Language: English
Subfield: Statistics and Probability
Field: Mathematics
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W2248861815