linear regression output as probabilities


Why shouldn’t you use linear regression outputs as probabilities?

in progress 0
Rohitsingh 5 months 1 Answer 279 views Member 0

Answer ( 1 )

  1. It’s tempting to use linear regression outputs as probabilities but it’s a mistake
    because the output can be negative, and greater than 1 whereas probability can not. As
    regression might actually produce probabilities that could be less than 0, or even bigger
    than 1, logistic regression was introduced.

Leave an answer