The Kepler mission, which has been in operation since 2009, is an ambitious space-based sur- vey of over 100,000 Milky Way stars with the goal of detecting planets transiting in front of those stars [1]. Already, the Kepler mission has proved extremely successful in finding transiting exo- planets; however, many stars in the survey are hard to analyze using the Kepler pipeline because the light emitted from the star is not at a very steady level. Gabriel Caceres and Eric Feigelson surmised that many of those stars displayed autoregressive behavior that could be modeled out, thereby increasing sensitivity to any dips in the light curves caused by planetary transits. I helped implement some autoregressive modeling to stars observed by Kepler and then used an algorithm devised by Caceres to search for planets in the light curves of stars with autoregressive behavior modeled out. I investigated the effects of increasing model complexity on the visibility of a tran- siting planet’s signal and found that models with only two or three parameters offered substantial improvement in a light curve’s variability without overmodeling the curve and obscuring the signal of a transiting planet.