The Sampling Distribution of Age-Specific Survival Estimates from an Age Distribution

Publisher Website
Google Scholar
Abstract
The sampling distributions for age-specific survival rates based on age distributions were simulated. The simulated sampling distributions tended to be both positively skewed and kurtotic. For a long-lived animal with high survival rates and for sample sizes up to 1000, there is a large chance that the survival rate estimate for at least 1 age-class will exceed 1. When survival rates are > 0.50, large oscillations can be expected around a smooth survival curve in the estimates of age-specific rates because of large negative correlation exceeding -0.41. The associated variances and covariances need to be considered in evaluations of age-specific survival estimates, in comparisons of survival curves, and in their application in population models.

This publication has 0 references indexed in Scilit: