Period Estimation and Denoising Families of Nonuniformly Sampled Time Series
William Seguine · Digital Commons - East Tennessee State University (East Tennessee State University) · 2019
Nonuniformly sampled time series are common in astronomy, finance, and other areas of research. Commonly, these time series belong to a family of signals recorded from the same phenomenon. Period estimation and denoising of such data relies on periodograms. In particular, the Lomb-Scargle periodogram and its extension, the Multiband Lomb-Scargle, are at the forefront of time series period estimation. However, these methods are not without laws. This paper explores alternatives to the Lomb-Scargle and Multiband Lomb-Scargle. In particular, this thesis uses regularized least squares and the convolution theorem to introduce a spectral consensus model of a family of nonuniformly sampled time series.