The sample value can't really be a negative number, as it's supposed to reflect the amount of nitrate in it. The blank is adjusted to 0, and the sample should have some absorbance greater than 0. Practically speaking, errors in measurement and impurities can give you results like yours. Ideally, it's supposed to work like this:
Blank = 0 adsorbance
Sample absorbance = some positive value "A" due to Nitrate ion in the sample
Control absorbance = some positve vaule "B", due to Nitrate in sample ("A") PLUS absorbance due to 0.01% Nitrate.
The difference between A and B should reflect the absorbance due to 0.01% nitrate, and A should be due only to the nitrate in the sample. Since you have interference or errors from someplace, giving a less than 0 absorbance for the sample, I'd run the test again; the -0.05 sample absorbance is a bit big. If you get similar results, but the sample absorbance is close to 0 (+ or - 0.02, maybe) just note that the nitrate concentration is below the detection limit of the method.
And no, you can't have a negative percent of Nitrate, except on philosophy classes.