Name: Leticia Vásquez
Institution: Universidad de Costa Rica
E-mail: letivg169@gmail.com
Co-authors:
Abstract:
Bayesian analysis of means comparison for independent samples was carried out through a study of child malnutrition in Costa Rica. A normal model was implemented assuming a previous distribution with known variance and a normal model assuming a previous distribution with unknown variance..
The Markov Chains via Monte Carlo (MCMC) method was used for the analysis of a normal model assuming a previous distribution with known variance. On the other hand, an analysis was made with the normal model assuming a previous distribution with both the mean and the unknown variance. For both models a diagnosis of convergence of the chains was made.
The main results it is observed that both the analysis with the normal model assuming a previous distribution with known variance using a normal priori and with the analysis with the normal model assuming a prior distribution with both the mean and the unknown variance, reached that to make the difference of means test in both cases the null hypothesis was rejected and in addition it was observed that in the chains created with the different models they converged, and it is concluded that the children present a higher degree of the malnutrition index than the girls.