I have read other similar questions, but all the answers seem to be something along the lines that without an intercept, you are ensuring that when the predictor is zero, the target variable is also zero, or how we need to add b to the equation y= mx+b, otherwise, we are forcing the regression line to go through the origin, which would bias our data.
My questions are, all statmodel add_constant do is create a column of ones, how does that creates the mathematical term b in y= mx+b
Second, even if we assumed b was added to the equation, would not that force the line to go through y=1 when x is zero? how is that so different from the line going through the origin.
