Javascript must be enabled for the correct page display

Evaluation of a General Bayesian Method using the Cholesky Decomposition to model the transmission of intelligence

Heeg, Dagmar (2019) Evaluation of a General Bayesian Method using the Cholesky Decomposition to model the transmission of intelligence. Research Project, Science Education and Communication.

[img]
Preview
Text
mSEC_2019_HeegDM.pdf

Download (2MB) | Preview
[img] Text
toestemming.pdf
Restricted to Registered users only

Download (120kB)

Abstract

The classical twin design has proven to be a powerful method in order to study the transmission of intelligence from parent to child. However, it comes with a few un- realistic assumptions such as random mating. In this paper the classical twin design is extended with cultural transmission, phenotypic assortment, social homogamy, the dominance effect or a combination of these extensions. In order to avoid analyzing sum scores, the Rasch Model is used as a measurement model and both the genetic and measurement model are put in a Bayesian framework, in order to estimate both simultaneously. A Cholesky decomposition is applied to make the models more easy to access and finally a Metropolis-Hasting algorithm was implemented to keep the model transparent. Six different models are compared by calculating the DIC scores, to find the genetic model that represents the real-world data the best.

Item Type: Thesis (Research Project)
Supervisor:
Supervisor nameSupervisor E mail
Grzegorczyk, M.A.M.A.Grzegorczyk@rug.nl
Degree programme: Science Education and Communication
Thesis type: Research Project
Language: English
Date Deposited: 20 Feb 2019
Last Modified: 04 Mar 2019 10:39
URI: http://fse.studenttheses.ub.rug.nl/id/eprint/19182

Actions (login required)

View Item View Item