Can anyone solve this question

[Amey12]

This question relates to a SAS Proc GLM analysis of data from a cross-sectional survey looking at the relationship between Y and D (which are quantitative variables). The table below summarises a sequence of fitted models with Y = dependent variable.

Which type of curvature improves the fit of the model the most?

D × ln(D)

sqrt(D)

ln(D)

D × D

[Ksharp]

Better post it at STAT Forum:
https://communities.sas.com/t5/Statistical-Procedures/bd-p/statistical_procedures
And I guess it is ln(D) ?

[Amey12]

Are you sure on this

[Ksharp]

Nope.
That is the reason why I ask you to post it at STAT Forum. And let statistical experts to see it . @StatDave @lvm @SteveDenham

[Amey12]

How do i post it on stat forum

[Ksharp]