Confusion concerning transition wavenumbers

I have been trying to calculate the wavenumber for the 2p -> 5d transition in hydrogren. Using the formula: $ R_H (frac{1}{n^2_f}-frac{1}{n^2_i}) $ would give me a negative answer. I have never seen a negative answer to this type of question before, but though it feels intuitively wrong I can’t think of why this would be an issue?