The Effect of Features on the Functional Form of the Scalar Power Spectrum

Abstract

We study how the scalar power spectrum of single-scalar inflation depends functionally on models with features which have been proposed to explain anomalies in the data. We exploit a new formalism based on evolving the norm-squared of the scalar mode functions, rather than the mode functions themselves.

27 pages, 14 figures, uses LaTeX2e; version 2 contains an expanded section 4 which explains how to compute the phase of the mode function so that the propagator and non-Gaussianity can be derived