Measuring the effect of individual attributes such as body mass upon animal survival rates
Capture-recapture methods are commonly used to study the population dynamics of animal populations. At each sampling occasion, animals are captured. If they are untagged, a tag is attached and the animal is released. If the animal was previously tagged, the tag number is recorded, and the animal is released. The Cormack-Jolly-Seber (CJS) model is the most commonly used method to estimate survival rates of the animals from the set of capture histories in the study. Many forces can potentially affect survival rates - these are generally called covariates. Prior to Simon's work, covariates could be individually based, but static (e.g. males and females could have different survival rates), or individually based, but discrete (e.g. survival rates could depend upon the area an animal migrates). His work allows for covariates (such as body mass) that vary among animals and change continuously over time.
This type of interdisciplinary work is a hallmark of our program in Applied Statistics at Simon Fraser University. For more information, please contact Simon Bonner (sbonner (at) stat.sfu.ca) or his supervisor Carl Schwarz (cschwarz (at) stat.sfu.ca).