Did you do a spectral fit, and it converged, but it left you unsatisfied with its quality? Think it could have done better, but you can't justify pushing the model any further because the χ² is apparently "good enough"? Then this recipe, https://iopscience.iop.org/article/10.3847/2515-5172/ad18b5, by @anshumanastro and me is what you are looking for. It lets you analyze trends in the residuals and control systematics, allowing you to access and use information that wasn't used in the χ² minimization.
Add comment