-1

is the intercept B0 in y = B0 + B1X1 + .... fit differently for every feature x1.

Is it different for every feature coefficient or the same for all feature coefficients and why so?

develarist
  • 3,009
  • 8
  • 31
dhruv tyagi
  • 1

1 Answers1

3

NO, there is only one intercept in the model, with only one value. It is not clear from where your misconception comes, but from the algebra $$ y_i=\beta_0 + \beta_1 x_{i1} + \beta_2 x_{i2} +\dotso +\epsilon_i $$ which is the multiple linear regression model, the constant term (intercept) $\beta_0$ has only one index $_0$, is only one symbol, and can only represent one number.

The different predictor variables $x$ have indices $_{i1}, _{i2}, \dotsc$ the last number $1,2,\dotsc$ indicating which predictor variable it is. This number does not occur with the intercept, so there is no connection.

@user20637 says in a comment The intercept will (often) change if a predictor is removed or added. This may be the source of the OPs misconception. If so, the following post will help you: Why is the intercept in multiple regression changing when including/excluding regressors?

kjetil b halvorsen
  • 63,378
  • 26
  • 142
  • 467
  • 1
    The intercept will (often) change if a predictor is removed or added. This may be the source of the OPs misconception. – user20637 Oct 21 '20 at 19:49
  • Can not edit since it only involves one word, but it should be "have" and not "has" in the first sentence of the last paragraph. +1 though. – shenflow Oct 21 '20 at 19:56