linear regression output as probabilities
Question
Why shouldn’t you use linear regression outputs as probabilities?
in progress
0
Machine Learning
4 years
1 Answer
882 views
Member 0
Answer ( 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.