UVM Theses and Dissertations
Format:
Print
Author:
Bakerman, Jordan Eli
Dept./Program:
Statistics Program
Year:
2012
Degree:
M.S.
Abstract:
The proceeding analysis serves to educate the reader on the interpretation and parameter estimation of mixed models when trials are conducted at multiple centers. First, the interpretations of generic results are explained for fixed effects models and the mixed model extension. Next, parameter estimation for normal mixed models and logistic regression is derived with emphasis on the Newton-Raphson algorithm. SAS output, modeling techniques, and the GLIMMIX procedure are also analyzed. Lastly, knowledge from beginning chapters will be applied to a data set considering possible predictors of mastectomy rates.