[R] in R when I get negative adjusted R^2 using "lm", what might be the problem?

[Michael]

And in the non-log case, all the previously significant coefficients
now became insignificant...

On Sun, Nov 9, 2008 at 5:36 PM, Michael <comtech.usa at gmail.com> wrote:
> This is a linear regression of Y onto factors...
>
> If I take log of Y, and regress onto the factors, I got:
>
> Multiple R-squared: 0.4023,     Adjusted R-squared: 0.2731
>
> If I don't take log of Y, and directly regress Y onto the factors, I got:
>
> Multiple R-squared: 0.1807,     Adjusted R-squared: -0.001112
>
> Is this negative adjusted R^2 a problem?
>
> What observation can I make here and what might be the problem?
>
> Thanks!
>