For the following model:

$$T = A_0 + A_1M+ A_2M^2 + A_3\ln(P) + A_4\ln(P)^2+A_5M\ln(P)$$


$T$ = Dependent variable

$P$ = Independent variable

$M$ = Independent variable

$A_0 \dots A_5$ = Coefficients of the regression.

I would like to know which class of regression model this equation corresponds to, when you use this kind of model, and how you find the parameters $A_0 \dots A_n$.

  • 6
    $\begingroup$ It's a polynomial regression model of degree two (or sometimes, a 'quadratic model') in two predictors, ($M$ and $\ln P$). You estimate the coefficients the same way as for any other multiple regression. $\endgroup$ – Glen_b -Reinstate Monica Jun 10 '13 at 13:27
  • $\begingroup$ hi Glen_b: by chance would you know why there is used the "Ln()"? $\endgroup$ – user26695 Jun 10 '13 at 17:46
  • $\begingroup$ Presumably because whoever built the model felt that whatever T was would be more nearly linearly related to lnP than to P; without context and subject knowledge, it's impossible for me to say much. $\endgroup$ – Glen_b -Reinstate Monica Jun 11 '13 at 0:55

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.