Logistic Regression Model

Logistic Regression Model

1. When to use it:

   • One nominal variable with 2 values and one measurement value
     • The nominal value is the dependent variable
     • The measurement variable is the independent variable

2. Goals

   • to see whether the probability of getting a particular value of the nominal variable is associated with the measurement variable
   • to predict the probability of getting a particular value of the nominal variable, given the measurement variable

3. Predicting future activity

   • Through using the logistic regression model, the user can determine whether or not the variables are related or not.
     • This is useful in healthcare because this model could potentially be used to prove malpractice
     • This could be used in biology to prove that maybe it wasn't the practitioners fault, but maybe an anatomical malfunction in the patient
     • In terms of a social system, the regression output gives us information about the overall significance or how well it is

4. The most useful result from a logistic regression is the adjusted odds ratios