Healthcare Analytics: Regression in R

Course details

Linear and logistic regression models can be created using R, the open-source statistical computing software. In this course, biotech expert and epidemiologist Monika Wahi uses the publicly available Behavioral Risk Factor Surveillance Survey (BRFSS) dataset to show you how to perform a forward stepwise modeling process. Monika shows you how to design your research by considering scientific plausibility selecting a hypothesis. Then, she takes you through the steps of preparing, developing, and finalizing both a linear regression model and a logistic regression model. She also shares techniques for how to interpret diagnostic plots, improve model fit, compare models, and more.

Learning objectives
Dealing with scientific plausibility
Selecting a hypothesis
Interpreting diagnostic plots
Working with indexes and model metadata
Working with quartiles and ranking
Making a working model
Improving model fit
Performing linear regression modeling
Performing logistic regression modeling
Performing forward stepwise regression
Estimating parameters
Interpreting an odds ratio
Adding odds ratios to models
Comparing nested models
Presenting and interpreting the final model

Concepts

0. Introduction

01 - Welcome to the course
02 - What you should know
03 - Introduction to the course
04 - Using the exercise files

1. Designing Your Research

05 - Scientific method review
06 - Using a cross-sectional approach
07 - Reviewing existing literature for ideas
08 - Dealing with scientific plausibility
09 - Selecting a linear regression hypothesis
10 - Selecting a logistic regression hypothesis
11 - Installing necessary packages

2. Preparing for Linear Regression

12 - Plots for checking assumptions in linear regression
13 - Interpreting diagnostic plots
14 - Categorization and transformation
15 - Indexes
16 - Quartiles
17 - Ranking
18 - Regression review
19 - Preparing to report results

3. Beginning Linear Regression Modeling

20 - Choices of modeling approaches
21 - Overview of modeling process
22 - Linear regression output
23 - Models 1 and 2
24 - Model metadata

4. Final Linear Regression Modeling

25 - Beginning Model 3
26 - Making a working Model 3
27 - Finalizing Model 3
28 - Looking at the final model
29 - Fishing and interaction
30 - Other strategies for improving model fit
31 - Defending the final model
32 - Presenting the final model

5. Preparing for Logistic Regression

33 - Analogies to linear regression process
34 - Parameter estimates in logistic regression
35 - Odds ratio interpretation
36 - Basic logistic code
37 - Forward stepwise regression - First two rounds
38 - Forward stepwise regression - Round 3

6. Developing the Logistic Regression Model

39 - Running Model 1
40 - Adding odds ratios to models
41 - Model metadata
42 - Forward stepwise - Round 2
43 - Forward stepwise - Round 3
44 - Using AIC to assess model fit
45 - When to compare nested models
46 - How to compare nested models
47 - Models 1 and 2 presentation
48 - Model 3 presentation
49 - Interpreting the final model

Conclusion

50 - Review of metadata
51 - Review of the process
52 - Next steps