Time series exploration in Python and MATLAB: Unevenly sampled data, parametric modeling, and periodograms
Abstract
This talk will open a Special Session, "Exploring time series data in high energy astrophysics," beginning with a brief survey of emerging software aiming to help astronomers explore and model diverse time series data. We will then describe work from our Time Series Explorer (TSE) project, producing new algorithms and software in Python and MATLAB for exploratory analysis and statistical modeling of time series data. This talk will focus on software for parametric modeling via least squares/maximum likelihood/Bayesian approaches, and some specialized time series tools for handling unevenly sampled and point process data. We will also discuss some persistent misunderstandings about periodograms and Fourier power spectra, which can play multiple roles in parametric and nonparametric analyses of astronomical time series. Additional talks in the session will cover modeling AGN time series with stochastic process models, and modeling periodic, quasiperiodic and aperiodic variability in multichannel spectro-temporal data from X-ray binaries using cospectra and other tools.
- Publication:
-
AAS/High Energy Astrophysics Division
- Pub Date:
- March 2019
- Bibcode:
- 2019HEAD...1711501L