Difficult# Assessing Residual Plots for Linearity

APSTAT-UKKLIP

Two variables, $x$ and $y$, were taken on $10$ subjects and two separate regression models were fit to the data.

Regression $I$ yielded the following equation and residual plot:

$$\text{predicted log:} \quad y = ax + b$$

...where $a$ and $b$ are constants.

Regression $II$ yielded the following equation and residual plot:

$$\text{predicted log:}\quad y = d + log cx$$

...where $c$ and $d$ are constants.