Statistical Reasoning for Public Health 2: Regression Methods

开始时间: 04/22/2022 持续时间: 8 weeks

所在平台: CourseraArchive 课程类别: 数学 大学或机构: Johns Hopkins University(约翰•霍普金斯大学) 授课老师： John McGready

课程主页: https://www.coursera.org/course/statreasoning2

课程评论：没有评论

第一个写评论        关注课程

课程详情

Regression methods are paramount for expanding the scope of analyses beyond the tools covered in Statistical Reasoning for Public Health 1.  In this course, a conceptual framework is employed to highlight the similarities between linear, logistic and Cox Proportion Hazards methods, while also focusing on the differences, in terms of usage and the interpretations of results from such models.  The course will start by laying out the details for all the regression approaches in the “simple” scenario, involving relating an outcome to single predictor.  After this overview of simple regression, confounding and effect-modification will be compared and contrasted, and adjusted and stratum-specific estimates will be introduced. Finally, multiple-regression models will then be debuted, and will be used to assess confounding and effect modification, produce adjusted and stratum-specific estimates, and to allow for better prediction of an outcome via the use of multiple predictors.  Linear spline models and propensity score methods for adjustment will also be briefly introduced.  All of the topics covered in this course will be reinforced with multiple examples from current and classical public health and medical studies.

课程大纲

• Simple regression models: linear, logistic, and Cox proportional hazards regression
• Confounding
• Effect Modification
• Multiple regression models (linear, logistic, Cox) for:
• Prediction with multiple factors
• Estimating adjusted associations
• Assessing and testing for effect modification
• Techniques for handling violations of regression linearity assumptions

课程评论(0条)