Skip to main content

140.753.01
Advanced Methods in Biostatistics III

Location
East Baltimore
Term
3rd Term
Department
Biostatistics
Credit(s)
3
Academic Year
2013 - 2014
Instruction Method
TBD
Class Time(s)
Tu, Th, 10:30 - 11:50am
Auditors Allowed
Yes, with instructor consent
Available to Undergraduate
No
Grading Restriction
Letter Grade or Pass/Fail
Contact Name
Frequency Schedule
Every Year
Prerequisite

140.751-752; Students must also register for 140.754

Description
Introduces the General Linear Model and Generalized Least Squares. Develops the Generalized Likelihood Ratio Test (GLRT) and connects it to the Gaussian Linear Model. Defines Fisher Information and Observed Information. Compares methods of simultaneous inference and multiple comparisons. Covers robust variance estimation. Compares optimal statistical weights to optimal policy weights, and missing data theory and practice. Develops consequences of departures from assumptions, efficiency and robustness trade-offs in the context of missing data and correlated responses. Identifies implications for design, and outlines basic experimental designs, choice of design and analysis, fixed and random effects, Introduces shrinkage estimates. Covers study designs that account for uncertainty in input parameters. Introduces sample reuse via the jackknife and adds to criteria to use in evaluating a procedure and how to identify when a new method or adaptation is needed.
Learning Objectives
Upon successfully completing this course, students will be able to:
  1. give examples of different types of data arising in public health studies
  2. Discuss differences and similarities between standard linear regression and models for discrete outcomes
  3. use modern statistical concepts such as generalized linear models for inference
  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
  7. expand the set of biostatistical models with quasi-likelihood, beta-binomial and log-linear models
  8. improve computational and analytic skills through analysis of simulated data sets