[Population Modeling] On Parameter Estimation

[Amiyaal Ilany]

Hi,

Here are my two cents about parameter estimation in my work.

I study social networks in wild mammals, with a recent extension of
developing general models of social networks. I estimated the contribution
of environmental, individual, and structural factors to social network
dynamics in a population of spotted hyenas in Kenya over 20 years:
http://onlinelibrary.wiley.com/doi/10.1111/ele.12447/abstract

I used the RSiena package to statistically estimate model parameters. The
general approach is using agent-based simulations to find the coefficients
of a model based on a multinomial logit function. Many details about
different approaches to estimate these models can be found in the package
manual: https://www.stats.ox.ac.uk/~snijders/siena/RSiena_Manual.pdf

In a second project, I developed (with Erol Akcay) a general model of
social network structure in animals. Our model is based on the idea of
social inheritance, where individuals "inherit" their social affiliations
from their parents. We fitted the model to data from observed wild animals
using partial least squares regression. Using this method we could estimate
the model parameters for different species. The manuscript describing this
work can be found here: http://biorxiv.org/content/early/2015/10/19/026120

All the best,
Amiyaal Ilany


-- 
Dr. Amiyaal Ilany
Postdoctoral Fellow
Department of Biology
University of Pennsylvania
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://simtk.org/pipermail/popmodwkgrpimag-news/attachments/20160114/7ab78136/attachment.html>