# 140.755.01 ADVANCED METHODS IN BIOSTATISTICS V

Department:
Term: 1st term
Credits: 4 credits
Course Instructor:
Description:

Reviews the extension of linear models to generalized linear models. Includes exponential family models, link functions, and over-dispersion. Also introduces models and inferential methods for polytomous outcomes. Describes extension of models to account for clustering using explicit modeling via mixed effects framework and generalized estimating equations (GEE). Introduces methods and models for regression with covariates subject to measurement error. Describes and implements advanced computational algorithms, such as Markov Chain Monte Carlo (MCMC) and expectation maximization (EM).

Old Learning Objective:

Upon completion of this course, students will be able to: 1) Give examples of different types of data arising in public health studies; 2) Use modern statistical concepts such as Generalized Linear Models for inference; 3) Describe models for polytomous outcomes; 4) Apply theoretical concepts to scientific data using R and WinBUGS software; 5) Conduct and interpret logistic, conditional logistic, and probit regression inference; 6) Extend models to account for clustering and correlation; 7) Introduce the mixed effects framework and describe its relationship to multilevel models; 8) Introduce models that account for measurement error in the covariates; 9) Provide new computational tools for complex models including Markov Chain Monte Carlo (MCMC) and Expectation Maximization (EM) algorithms; 10) Improve computational and analytic skills through analysis of simulated data sets.

New Learning Objective(s):
Upon successfully completing this course, students will be able to:
Give examples of different types of data arising in public health studies
Use modern statistical concepts such as Generalized Linear Models for inference
Describe models for polytomous outcomes
Apply theoretical concepts to scientific data using R and WinBUGS software
Conduct and interpret logistic, conditional logistic, and probit regression inference
Extend models to account for clustering and correlation
Introduce the mixed effects framework and describe its relationship to multilevel models
Introduce models that account for measurement error in the covariates
Provide new computational tools for complex models including Markov Chain Monte Carlo (MCMC) and Expectation Maximization (EM) algorithms
Improve computational and analytic skills through analysis of simulated data sets

Methods of Assessment: Homeworks and a final exam
Location: East Baltimore
Class Times:
• Tuesday 10:30 - 11:50
• Thursday 10:30 - 11:50
Lab Times:
• Tu Th 10:00 - 10:30 (1)
Enrollment Minimum: 10
Instructor Consent: No consent required
Prerequisite:

140.751-4

Auditors Allowed: Yes, with instructor consent