# Can I use log and without any transformation variables in the one model for the independent variables?

Can I use log-log/linear regression models by taking the natural logarithm of some variables?

Imagine I have a dependent variable $$y$$ and independent variables $$x_1$$, $$x_2$$, and $$x_3$$. Can I run the following regression:

$$\log(y) = \log(x_1)+\log(x_2) + x_3$$