An implementation of Bayesian lensing shear measurement

The Bayesian gravitational shear estimation algorithm developed by Bernstein and Armstrong (2014) can potentially be used to overcome multiplicative noise bias and recover shear using very low signal-to-noise ratio (S/N) galaxy images. In that work the authors confirmed the method is nearly unbiased in a simplified demonstration, but no test was performed on images with realistic pixel noise. Here I present a full implementation for fitting models to galaxy images, including the effects of a point spread function (PSF) and pixelization. I tested the implementation using simulated galaxy images modeled as Sersic profiles with n=1 (exponential) and n=4 (De Vaucouleurs'), convolved with a PSF and a flat pixel response function. I used a round Gaussian model for the PSF to avoid potential PSF-fitting errors. I simulated galaxies with mean observed, post-PSF full-width at half maximum equal to approximately 1.2 times that of the PSF, with log-normal scatter. I also drew fluxes from a log-normal distribution. I produced independent simulations, each with pixel noise tuned to produce different mean S/N ranging from 10-1000. I applied a constant shear to all images. I fit the simulated images to a model with the true Sersic index to avoid modeling biases. I recovered the input shear with fractional error less than 2 x 10^{-3} in all cases. In these controlled conditions, and in the absence of other multiplicative errors, this implementation is sufficiently unbiased for current surveys and approaches the requirements for planned surveys.