Parameter uncertainties in weighted unbinned maximum likelihood fits

Abstract Parameter estimation via unbinned maximum likelihood fits is central for many analyses performed in high energy physics. Unbinned using event weights, example to statistically subtract background contributions the sPlot formalism, or correct acceptance effects, have recently seen increasing use community. However, it well known that naive approach of parameter uncertainties second derivative logarithmic does not yield confidence intervals with coverage presence weights. This paper derives asymptotically expressions and compares them several commonly used approaches determination uncertainties, some which are shown generally be correct. In addition, effect on weights discussed, including can arise from nuisance parameters sWeights .