Looking at this the functional form "quadratic model" is introduced as a linear regression model does not assume linearity between IV and DV and only has to be linear in parameters. Is it still a linear model when fitted with OLS?
Asked
Active
Viewed 21 times
0
-
2See here https://stats.stackexchange.com/a/597445/56940 – utobi Jan 05 '23 at 17:49
-
As a heads up, a model is linear regardless of estimation approach (e.g., OLS vs minimizing the sum of the absolute value of the residuals). Indeed, a model is linear whether you estimate the coefficients or not! – Dave Jan 05 '23 at 17:58
-
Curious even when one introduces X^2 the parameters still enter the model linear don't they. TBH I do not really agree with the accepted answer you suggested @utobi – cs0815 Jan 05 '23 at 19:14
-
@cs0815 could you explain why you disagree? – utobi Jan 05 '23 at 19:29
-
X^2 is feature engineered - the parameters therefore still enter the model linearly making it a linear model? – cs0815 Jan 06 '23 at 08:12