Corequisites: MPBH 421 (or prior completion of a two-semester course sequence in Biostatistics) and one course in statistical computing such as MPBH 412.
This course covers a broad overview of statistical models and estimation methods for outcome variables (normal and non-normal) that are clustered or measured repeatedly in time or space. The focus is on applications and computer software methods for correlated regression models, including ANOVA based methods, hierarchical linear models, etc.
Outcomes: Apply generalized linear models to regress a set of explanatory variables against non-normally distributed outcomes, including binary outcomes (logistic regression), multinomial outcomes (ordinal and multinomial logistic regression), and count outcomes (Poisson and negative-binomial regression); Identify and employ appropriate methods for the analysis of repeated measures, including the use of generalized estimating equations (GEE models) and/or mixed-effects models for nested and hierarchical data; Analyze data using Parametric methods for time-to-event data; Conduct Bayesian estimation for unknown parameters; Employ multivariate methods to solve real-world problems and dimensionality reduction; Use statistical programming to apply learned statistical models and estimation methods; Explain the results of statistical analyses.
This course covers a broad overview of statistical models and estimation methods for outcome variables (normal and non-normal) that are clustered or measured repeatedly in time or space. The focus is on applications and computer software methods for correlated regression models, including ANOVA based methods, hierarchical linear models, etc.
Outcomes: Apply generalized linear models to regress a set of explanatory variables against non-normally distributed outcomes, including binary outcomes (logistic regression), multinomial outcomes (ordinal and multinomial logistic regression), and count outcomes (Poisson and negative-binomial regression); Identify and employ appropriate methods for the analysis of repeated measures, including the use of generalized estimating equations (GEE models) and/or mixed-effects models for nested and hierarchical data; Analyze data using Parametric methods for time-to-event data; Conduct Bayesian estimation for unknown parameters; Employ multivariate methods to solve real-world problems and dimensionality reduction; Use statistical programming to apply learned statistical models and estimation methods; Explain the results of statistical analyses.