Bayesian PSD estimation for LISA noise based on P-splines with a parametric boost

This repository provides datasets to evaluate Bayesian power spectral density (PSD) estimation for LISA X-channel noise using P-splines with a parametric component. The contents are: X channel noise-4a (LISA SGWB Dataset (noise-4a)) LISA dataset periodograms and results (durations: 3, 6 months, and 1 year): ddpc_f_3mon.txt, ddpc_f_6mon.txt, ddpc_f_1year.txt: Fourier frequencies ddpc_x2_per_3mon.txt, ddpc_x2_per_6mon.txt, ddpc_x2_per_1year.txt: Blocked periodogram. mcmc_results_1year.h5, mcmc_results_1year.h5, mcmc_results_1year.h5: MCMC outputs and posterior summaries. AR(4) simulations and results. Each archive contains 500 instances of periodograms, parametric model PSDs, MCMC results of Model 1 (when the parametric model is white noise), and Model 2 (when the parametric model is AR(4)). arsim_128.7z arsim_256.7z arsim_512.7z Integrate Absolute error files: iae_ar0.txt, iae_ar4_knots.txt: IAE of 1500 AR(4) simulations using Model 1 and 2. iae_ar0_knots.txt, iae_ar4_knots.txt: IAE of 150 AR(4) simulations using Model 1 and 2 for various knots allocations.