Autoregressive Times Series Methods for Time Domain Astronomy

Frontiers in Physics

Caceres, Gabriel A.¹  Babu, G. Jogesh¹  Feigelson, Eric D.² 

[1] Center for Astrostatistics, Pennsylvania State University, United States;Department of Astronomy and Astrophysics, Pennsylvania State University, United States

关键词: time domain astronomy;    irregularly sampled time series;    variable stars;    quasars;    statistical methods;    time series analysis;    autoregressive modeling;    ARIMA;    CARMA;   

DOI  :  10.3389/fphy.2018.00080

学科分类：物理（综合）

来源: Frontiers

【 摘 要 】

Celestial objects exhibit a wide range of variability in brightness at different wavebands. Surprisingly, the most common methods for characterizing time series in statistics - parametric autoregressive modeling - is rarely used to interpret astronomical light curves. We review standard ARMA, ARIMA and ARFIMA (autoregressive moving average fractionally integrated) models that treat short-memory autocorrelation, long-memory 1/f^a `red noise', and nonstationary trends. Though designed for evenly spaced time series, moderately irregular cadences can be treated as evenly-spaced time series with missing data. Fitting algorithms are efficient and software implementations are widely available. We apply ARIMA models to light curves of four variable stars, discussing their effectiveness for different temporal characteristics. A variety of extensions to ARIMA are outlined, with emphasis on recently developed continuous-time models like CARMA and CARFIMA designed for irregularly spaced time series. Strengths and weakness of ARIMA-type modeling for astronomical data analysis and astrophysical insights are reviewed.

【 授权许可】

CC BY