Using M-type smoothing splines to estimate the spectral density of a stationary time series

✍ Scribed by Eva Ferreira

• Publisher: Elsevier Science
• Year: 1998
• Tongue: English
• Weight: 507 KB
• Volume: 38
• Category: Article
• ISSN: 0167-7152
• DOI: 10.1016/s0167-7152(98)00011-x

✦ Synopsis

An important problem in time series is the estimation of the spectral density or spectrum of a stationary process. In this paper we propose to use an M-type smoothing spline to estimate the spectrum. We derive some general results for these type of estimators which allow us to choose the optimal M-type spline in some asymptotical sense. Applying the theoretical results to our objective, the estimation of the spectrum, the selected estimator improves the classical estimation with least-squares splines. (~