A suite of tools for denoising one-dimensional signals with trend filtering. Optimizing the hyparameter via Stein's unbiased risk estimate or V-fold cross validation with a customizable validation loss functional, and one-standard-error rules. Relaxed trend filtering, and various bootstrap algorithms for uncertainty quantification.

## Installation

install.packages("remotes")
remotes::install_github("capolitsch/trendfiltering")

## Usage terms

When using the trendfiltering package, please use the following citations in your publications:

1. Politsch et al. (2020a). Trend filtering – I. A modern statistical tool for time-domain astronomy and astronomical spectroscopy. Monthly Notices of the Royal Astronomical Society, 492(3), p. 4005-4018. [URL] [BibTeX]

2. Politsch et al. (2020b). Trend Filtering – II. Denoising astronomical signals with varying degrees of smoothness. Monthly Notices of the Royal Astronomical Society, 492(3), p. 4019-4032. [URL] [BibTeX]