Propensity Score
​
The Propensity Score matching technique is used to make observational data "similar" to randomized data. It is useful for numerous cases and with numerous explanatory variables.
​
​Objectives:
Using the logistic regression model, estimate the parameters that explain the variability of dichotomous or multinomial dependent variables. The estimated parameters are odds ratios.
Techniques presented:
Contingency tables, risk coefficients (relative risk, odds ratio), logistic regression, multinomial regression.
​
Tutorials:
There are exercises for each of the topics covered.
Prerequisites:
Attending the TSC course or having knowledge of the topics it contains is preparatory.
It is advisable to attend the REG course.
Subjects:
Contingency tables
The risk indices:
-
Relative risk
-
odds ratio
-
Hazard Ratio
The logistic regression model:
-
Binomial
-
Multinomial
The propensity score:
-
Its purposes
-
The different ways of application
-
The different types of matching
-
How to analyze matched data