Title “Nonparametric reconstruction of the cosmic expansion with local regression smoothing and simulation extrapolation”
Abstract: Because today more than ever the cosmological data sets are getting even larger and it is necessary to extract accurate predictions from them, I will present a very brief review on the standard approach used to infer the best possible constraints from data. This kind of restrictions is part of the stream called parametric analysis. However, there is another stream, a nonparametric approach, in which a cosmological model is not needed. I will show how instead of standard observational tests, the global trend is inferred directly from raw data through the reconstruction based on a locally weighted scatterplot method and a simulation-extrapolation method. It is worth noting that the strength of this approach relies on its model-independent and nonparametric method, as well as does not assume any prior or energy content in the Universe nor any other property related to a cosmological model, apart from homogeneity and isotropy.