What is it about?
In many empirical applications we get a partial sample from a population whose unknown size we want to estimate. The sample is built up by using, repeatedly, siome identification mechanism. Therefore, each unit of the population has associated the count of times he/she/it has been registered, and only units with count greater than zero enters the observed sample.
Featured Image
Why is it important?
We use the truncated count distribution under heterogenity of the probability of identification across units in the population to get an estimate of the population size. The estimate is based on a very simple regression model which can be easily programmed in any standard worksheet.
Perspectives
While computer code is available from the authors, the paper gives explicit formulas to build the model, to get the estimate and to calulate the (asymptotic) variance of the same. Therefore, any result in the paper can be easily reproduced, and the proposed approach can be simply used in similar empirical situations by practitioners, without the need for substantial computer skill.
Professor Marco Alfo'
Universita degli Studi di Roma La Sapienza
Read the Original
This page is a summary of: A flexible ratio regression approach for zero-truncated capture-recapture counts, Biometrics, February 2016, Wiley,
DOI: 10.1111/biom.12485.
You can read the full text:
Contributors
The following have contributed to this page
