In logistic regression, the log odds is modeled as a linear functions of the
-variables. Thus, the logistic regression equation is:
Once the logistic regression has been “run,” the software will calculate estimates for , , , … , which are usually denoted by , , , … , and read “alpha hat,” “beta hat sub 1,” and so one. Plugging in these estimates yields the fitted logistic regression equation:
- “log” refers to the natural logarithm (log to the base ) as is discussed further here.
- “odds” are discussed further here.
- For a basic introduction to logistic regression analysis, go to the home page of this web site (click here).