Adaptive Spectral Estimation for Nonstationary Time Series
Duration: 39 mins 12 secs
Share this media item:
Embed this media item:
Embed this media item:
About this item
| Description: |
Stoffer, D (University of Pittsburgh)
Friday 17 January 2014, 11:30-12:15 |
|---|
| Created: | 2014-01-24 11:51 |
|---|---|
| Collection: | Inference for Change-Point and Related Processes |
| Publisher: | Isaac Newton Institute |
| Copyright: | Stoffer, D |
| Language: | eng (English) |
| Abstract: | We propose a method for analyzing possibly nonstationary time series by adaptively dividing the time series into an unknown but finite number of segments and estimating the corresponding local spectra by smoothing splines. The model is formulated in a Bayesian framework, and the estimation relies on reversible jump Markov chain Monte Carlo (RJMCMC) methods. For a given segmentation of the time series, the likelihood function is approximated via a product of local Whittle likelihoods. The number and lengths of the segments are assumed unknown and may change from one MCMC iteration to another. |
|---|---|
Available Formats
| Format | Quality | Bitrate | Size | |||
|---|---|---|---|---|---|---|
| MPEG-4 Video | 640x360 | 1.94 Mbits/sec | 570.70 MB | View | Download | |
| WebM | 640x360 | 653.55 kbits/sec | 187.64 MB | View | Download | |
| iPod Video | 480x270 | 521.47 kbits/sec | 149.72 MB | View | Download | |
| MP3 | 44100 Hz | 249.78 kbits/sec | 71.78 MB | Listen | Download | |
| Auto * | (Allows browser to choose a format it supports) | |||||