Negative R-squared?

When I enter all my predictors into my model, it comes out with an adjusted R2 thats negative - what does this mean? I thought R2 could only be positive, as it's RxR!

Thanks for any advice!


TS Contributor
Are you using Excel?Microsoft delivers it with a flaw...Also, it's not a problem if there is no intercept in the model

Take into consideration that as the number of predictors approaches the number of observations this is likely to happen (more of an an "abrupt" memory, I can't cite this)
Ahh, I see. Phew! I don't think theres an intercept. I've entered all my (5) predictors using the "enter" command in SPSS in the same block.

Thanks :)