Title: A Semiparametric Approach to Incorporating Systematic Uncertainties into Bayesian X-ray Spectral Fitting Authors: Hyunsook Lee, Vinay L. Kashyap, Jeremy J. Drake, Alanna Connors, Rima Izem, Taeyoung Park, Pete Ratzlaff, Aneta Siemiginowska, David A. van Dyk, Andreas Zezas Abstract: We develop a unique methodology to incorporate systematic uncertainties into X-ray spectral fitting analysis. These uncertainties have been ignored in calibrating noisy astronomical data and as a consequence, error bars of interesting parameters are generally underestimated. Our strategy combines parametric Bayesian spectral fitting and nonparametric approximation of the detector characteristics, the source of systematic uncertainties. We describe our implementation of this method here, in the context of recently codified \chandra\ effective area uncertainties. We estimate the posterior probability densities of absorbed power-law model parameters that include the effects of systematic uncertainties. We apply our method to both simulated as well as actual \chandra\ ACIS-S data. Because of the modular structure of the Bayesian spectral fitting technique, incorporating such uncertainties can be executed simultaneously within the Markov chain Monte Carlo method. Therefore, our strategy itself does not significantly affect the overall computing time but offers adequate parameter estimates and error bars. This research was supported by NASA/AISRP Grant NNG06GF17G.