Intercept term in logistic regression

Suppose we have the following logistic regression model:

logit(p)=β0+β1x1+β2x2

Is β0 the odds of the event when x1=0 and x2=0? In other words, it is the odds of the event when x1 and x2 are at the lowest levels (even if this is not 0)? For example, if x1 and x2 take only the values 2 and 3 then we cannot set them to 0.

Answer

β0 is not the odds of the event when x1=x2=0, it is the log of the odds. In addition, it is the log odds only when x1=x2=0, not when they are at their lowest non-zero values.

Attribution
Source : Link , Question Author : logisticgu , Answer Author : gung – Reinstate Monica

Leave a Comment