{ "id": "epj6prkr3f", "slug": "multitaper", "title": "Multitaper", "updated": "2026-10-21", "topic_path": [ { "id": "physical", "label": "Physical world and mathematics", "api_url": "https://www.edgepedia.org/api/v1/topics/physical" }, { "id": "physical.math", "label": "Mathematics and statistics", "api_url": "https://www.edgepedia.org/api/v1/topics/physical.math" }, { "id": "physical.math.statistics", "label": "Statistics and probability", "api_url": "https://www.edgepedia.org/api/v1/topics/physical.math.statistics" }, { "id": "physical.math.statistics.stats_statistical_inference", "label": "Statistical inference, estimation, sampling, and testing", "api_url": "https://www.edgepedia.org/api/v1/topics/physical.math.statistics.stats_statistical_inference" }, { "id": "physical.math.statistics.stats_statistical_inference.estimation", "label": "Estimation theory and estimator families", "api_url": "https://www.edgepedia.org/api/v1/topics/physical.math.statistics.stats_statistical_inference.estimation" } ], "geo": [ { "id": "geo.nongeo.t1946.physical.math.statistics.stats_statistical_inference.estimation", "label": "Non-geographic · 1946 to 2000: Estimation theory and estimator families", "api_url": "https://www.edgepedia.org/api/v1/geo/geo.nongeo.t1946.physical.math.statistics.stats_statistical_inference.estimation", "path": [ { "id": "geo.nongeo", "label": "Non-geographic", "api_url": "https://www.edgepedia.org/api/v1/geo/geo.nongeo" }, { "id": "geo.nongeo.t1946", "label": "Non-geographic · 1946 to 2000", "api_url": "https://www.edgepedia.org/api/v1/geo/geo.nongeo.t1946" }, { "id": "geo.nongeo.t1946.physical", "label": "Physical world and mathematics", "api_url": "https://www.edgepedia.org/api/v1/geo/geo.nongeo.t1946.physical" }, { "id": "geo.nongeo.t1946.physical.math", "label": "Mathematics and statistics", "api_url": "https://www.edgepedia.org/api/v1/geo/geo.nongeo.t1946.physical.math" }, { "id": "geo.nongeo.t1946.physical.math.statistics", "label": "Statistics and probability", "api_url": "https://www.edgepedia.org/api/v1/geo/geo.nongeo.t1946.physical.math.statistics" }, { "id": "geo.nongeo.t1946.physical.math.statistics.stats_statistical_inference", "label": "Statistical inference, estimation, sampling, and testing", "api_url": "https://www.edgepedia.org/api/v1/geo/geo.nongeo.t1946.physical.math.statistics.stats_statistical_inference" }, { "id": "geo.nongeo.t1946.physical.math.statistics.stats_statistical_inference.estimation", "label": "Estimation theory and estimator families", "api_url": "https://www.edgepedia.org/api/v1/geo/geo.nongeo.t1946.physical.math.statistics.stats_statistical_inference.estimation" } ] } ], "excerpt": "The multitaper method is a nonparametric spectral estimation technique that reduces variance by averaging periodograms from orthogonal Slepian tapers, reported by D. J. Thomson in 1982.", "snippet": "The multitaper method is a nonparametric spectral estimation technique that reduces variance by averaging periodograms from orthogonal Slepian tapers, reported by D. J. Thomson in 1982.", "node": "physical.math.statistics.stats_statistical_inference.estimation", "markdown": "# Multitaper\n\nThe multitaper method is a nonparametric spectral estimation technique that reduces the variance of a power spectrum estimate by averaging the periodograms (estimate of a signal's power across frequencies) obtained after multiplying a time series by several orthogonal data tapers, most often Slepian (discrete prolate spheroidal sequence, DPSS) tapers.[1](https://arxiv.org/html/2103.11586v1) Averaging K tapered periodograms is more robust than relying on a single tapered periodogram, and the spectral concentration properties of the Slepian tapers mitigate spectral leakage.[1](https://arxiv.org/html/2103.11586v1) By applying K tapers to the entire record and averaging the resulting eigenspectra, the method gains multiple spectral degrees of freedom while keeping the frequency resolution of the full series.[1](https://arxiv.org/html/2103.11586v1) • [2](https://doi.org/10.1109/proc.1982.12433) D. J. Thomson reported the method in 1982, and it is now applied across radar, seismic, climate, and EEG analysis.[2](https://doi.org/10.1109/proc.1982.12433) • [3](https://www.embs.org/tbme/articles/review-multitaper-spectral-analysis/)\n\n| Key fact | Detail |\n|---|---|\n| What it estimates | The power spectral density, as an average of K tapered periodograms with \\( K \\approx 2 \\cdot N \\cdot W \\)[1](https://arxiv.org/html/2103.11586v1) |\n| Tapers | Orthogonal DPSS sequences, each maximizing spectral concentration in a band of half-width W[4](https://research.atmos.ucla.edu/tcd/ssa/guide/mann/mann4.html) • [5](https://pyspectrum.readthedocs.io/en/latest/_modules/spectrum/mtm.html) |\n| Taper-count rule | \\( K < 2 \\cdot p - 1 \\) for usefully small leakage[4](https://research.atmos.ucla.edu/tcd/ssa/guide/mann/mann4.html); in practice \\( K \\leq 2 \\cdot N \\cdot W - 1 \\), since too many tapers bias the estimate badly[6](https://iopscience.iop.org/article/10.3847/1538-3881/adc9b4/meta) |\n| Statistics | For locally white noise the high-resolution spectrum is chi-squared distributed with 2K degrees of freedom[4](https://research.atmos.ucla.edu/tcd/ssa/guide/mann/mann4.html) |\n| Typical parameters | NW = 4 is the common software default (R multitaper: nw = 4.0, k = 7)[7](https://cran.r-project.org/web/packages/multitaper/multitaper.pdf) • [8](https://www.osti.gov/pages/servlets/purl/1402465); climate records often use K = 3, p = 2[4](https://research.atmos.ucla.edu/tcd/ssa/guide/mann/mann4.html) |\n| Origin | D. J. Thomson, \"Spectrum estimation and harmonic analysis,\" Proceedings of the IEEE, 1982[2](https://doi.org/10.1109/proc.1982.12433) |\n| Software | R multitaper package: adaptive estimation, coherence, harmonic F-test, complex demodulation, jackknifed 95% confidence intervals[7](https://cran.r-project.org/web/packages/multitaper/multitaper.pdf) |\n\n## How it works\n\nA taper is a sequence of weights that multiplies the data before a [Fourier transform](https://www.edgepedia.org/fourier-transform). Power from strong spectral features can leak into frequencies where the true spectrum is small; this is spectral leakage, which the spectral concentration properties of the Slepian tapers mitigate.[1](https://arxiv.org/html/2103.11586v1) The DPSS tapers are chosen to maximize the fraction of the taper's energy concentrated in the frequency band (f − W, f + W). The second Slepian sequence maximizes that concentration ratio subject to being orthogonal to the first, the third is orthogonal to both, and so on.[5](https://pyspectrum.readthedocs.io/en/latest/_modules/spectrum/mtm.html)\n\nOrthogonality is what makes the averaging work: the K eigenspectra carry distinct information about the band, so averaging them reduces variance instead of duplicating one estimate.[4](https://research.atmos.ucla.edu/tcd/ssa/guide/mann/mann4.html) The time-bandwidth parameter NW sets the half-bandwidth \\( 2 \\cdot N \\cdot W \\) and hence the resolution-variance tradeoff: only the first 2p − 1 tapers have usefully small leakage, so K must stay below \\( 2 \\cdot p - 1 \\).[4](https://research.atmos.ucla.edu/tcd/ssa/guide/mann/mann4.html) If the power spectral density S(f) is twice differentiable, choosing \\( W = O(N^{-1/5}) \\) and \\( K \\approx 2 \\cdot N \\cdot W = O(N^{4/5}) \\) tapers minimizes the mean squared error of the estimate.[1](https://arxiv.org/html/2103.11586v1) For a white or locally white noise process the high-resolution spectrum is chi-squared distributed with \\( 2 \\cdot K \\) degrees of freedom; adaptive weighting reweights the K eigenspectra to guard against broadband leakage when the spectrum is colored but locally white.[4](https://research.atmos.ucla.edu/tcd/ssa/guide/mann/mann4.html)\n\n## How it is done\n\n1. Choose the time-bandwidth parameter NW and the number of tapers K, with \\( K \\leq 2 \\cdot N \\cdot W - 1 \\); using too many tapers leads to badly biased estimates.[6](https://iopscience.iop.org/article/10.3847/1538-3881/adc9b4/meta) Common defaults are NW = 4 with K = 7 in the R multitaper package; in general, K is typically set near \\( 2 \\cdot N \\cdot W - 1 \\).[7](https://cran.r-project.org/web/packages/multitaper/multitaper.pdf)\n2. Compute the DPSS tapers by solving a tridiagonal eigensystem whose diagonal elements are \\( \\frac{N-1-2n}{2}\\cos(2\\pi p/N) \\) and whose off-diagonal elements are \\( \\frac{n(N-n)}{2} \\), where the time-bandwidth product p scales the problem.[9](https://www.leesj.sites.oasis.unc.edu/FETCH/GRAB/Papers/leesandParkMTM.pdf)\n3. Form the eigencoefficients \\( J_{k}(f) = \\sum_{n=1}^{N} x(n)\\, v_{n}^{(k)}(N, f_{w})\\, e^{-j 2\\pi f n / f_{s}} \\), where \\( v_{n}^{(k)} \\) is the kth taper and \\( f_{s} \\) the sampling frequency, and compute the kth eigenspectrum as \\( |J_{k}(f)|^{2} \\).[10](https://www.sciencedirect.com/science/article/abs/pii/S1051200425008504)\n4. Combine the eigenspectra, either as the simple high-resolution average or with adaptive weights that downweight higher-order eigenspectra when leakage contributes noise to the band.[4](https://research.atmos.ucla.edu/tcd/ssa/guide/mann/mann4.html)\n5. Optionally run the harmonic F-test for line components.\n\nFor typical instrumental climate records, K = 3 with p = 2 offers a compromise between the frequency resolution needed to resolve distinct climate signals and the benefit of multiple spectral degrees of freedom; the optimal choice is application specific.[4](https://research.atmos.ucla.edu/tcd/ssa/guide/mann/mann4.html)\n\n## Origin\n\nThe method resulted from work by D. J. Thomson at [Bell Labs](https://www.edgepedia.org/bell-labs) in the 1970s and 1980s, culminating in the 1982 Proceedings of the IEEE paper \"Spectrum estimation and harmonic analysis.\"[2](https://doi.org/10.1109/proc.1982.12433) • [11](https://www.birs.ca/workshops/2022/22w2230/report22w2230.pdf) In that paper, contributions to the kth spectrum estimate from the region (f − W, f + W) are treated as \"signal\" and the rest of the frequency band as \"noise,\" with higher-order eigenspectra, which have poorer leakage resistance, downweighted accordingly, especially where the spectrum changes rapidly.[12](https://www.math.ucdavis.edu/~saito/data/ONR15/thomson_spect-est-harm-anal.pdf) • [20](https://pmc.ncbi.nlm.nih.gov/articles/PMC4179109/) The mathematics rests on the prolate spheroidal wave functions; Thomson's paper builds on what it calls a remarkable series of Slepian papers.[11](https://www.birs.ca/workshops/2022/22w2230/report22w2230.pdf) • [12](https://www.math.ucdavis.edu/~saito/data/ONR15/thomson_spect-est-harm-anal.pdf) Averaging periodograms of data segments was established earlier in Welch's method, but the multitaper method averages tapered transforms of the entire record, which preserves frequency resolution.[3](https://www.embs.org/tbme/articles/review-multitaper-spectral-analysis/) • [13](https://arxiv.org/html/2209.15027)\n\n## Variants\n\n**Adaptive weighting** reweights the K eigenspectra so that broadband leakage into a colored but locally white spectrum does not dominate the average.[4](https://research.atmos.ucla.edu/tcd/ssa/guide/mann/mann4.html) **The harmonic F-test** tests the ratio of the variance captured by the filtered portion of the series, using K eigentapers, to the residual variance with a Fisher-Snedecor F-test; it can detect low-amplitude harmonic oscillations in relatively short series with high statistical significance.[4](https://research.atmos.ucla.edu/tcd/ssa/guide/mann/mann4.html) In **spectral reshaping**, the effect of statistically significant lines is subtracted from the eigencomponents \\( Y_{k} \\) before adaptive weighting, allowing quadratic multitaper (QMT) estimates of the remaining stochastic part of the spectrum.[14](https://fanchung.ucsd.edu/ron/papers/07_08_multitaper.pdf)\n\n**Sine tapers** replace the DPSS eigensystem with analytically defined sinusoidal tapers: they achieve smaller local bias than Slepian tapers at the expense of sidelobe suppression, and K sine tapers give a spectral window concentrated in \\( [-W_{0}, W_{0}] \\) with \\( W_{0} = (K+1)/(2(N+1)) \\).[15](http://staff.washington.edu/dbp/PDFFILES/mseplp.pdf) An enhanced adaptive sine multitaper algorithm has been reported to improve the accuracy and efficiency of power [spectral density estimation](https://www.edgepedia.org/spectral-density-estimation) for evaluating low-frequency gravitational-wave detection systems.[16](https://www.mdpi.com/2076-3417/15/7/3919) For nonuniformly sampled time series, the mtNUFFT periodogram, a multitaper nonuniform fast Fourier transform estimator, improves on the statistical issues of the Lomb-Scargle periodogram while providing a factor of 3 speedup in some applications.[17](https://iopscience.iop.org/article/10.3847/1538-3881/ad7029/pdf) • [6](https://iopscience.iop.org/article/10.3847/1538-3881/adc9b4/meta) An extension adds a multitaper harmonic F-test to mtNUFFT that detects strictly periodic signals in noise and estimates their frequencies, demonstrated on Kepler-91 asteroseismic data.[6](https://iopscience.iop.org/article/10.3847/1538-3881/adc9b4/meta) A computationally efficient suboptimal variant, M2NuFFT, targets fast exploration of nonuniformly sampled series.[10](https://www.sciencedirect.com/science/article/abs/pii/S1051200425008504)\n\n## Applications\n\nNonparametric spectral estimation with multitaper appears in applications from radar and seismic data analysis to electroencephalography and speech processing, including EEG analyses of anesthesia and sleep.[3](https://www.embs.org/tbme/articles/review-multitaper-spectral-analysis/) In geophysics, an early application of eigentaper spectral analysis used the bandwidth parameter \\( 2 \\cdot W = 2 \\cdot P / N \\),[18](http://www.jmlilly.net/papers/park87a-jgr.pdf) and multitaper spectral analysis has been shown to better detect the presence or absence of 60 Hz power-line pick-up in Canadian dynamite data and to characterize seismic exploration data more accurately in terms of coherence and signal-to-noise.[19](https://onlinelibrary.wiley.com/doi/10.1002/env.3170050308) In paleoclimate work, the SSA-MTM toolkit provides spectral estimation and signal reconstruction, and a red-noise generalization of the harmonic test assesses narrowband signals against a robust red-noise background, with significance levels from chi-squared quantiles with \\( 2 \\cdot K \\) degrees of freedom.[4](https://research.atmos.ucla.edu/tcd/ssa/guide/mann/mann4.html) In asteroseismology, the harmonic F-test has been applied to Kepler-91 data for precise frequency estimation of stellar oscillations.[6](https://iopscience.iop.org/article/10.3847/1538-3881/adc9b4/meta)\n\n## Limitations and alternatives\n\nThe method assumes a locally white background. [Colored noise](https://www.edgepedia.org/colors-of-noise) or chaotic systems can be broken into spurious lines with high F-values, so harmonic significance must be judged against an estimated noise background.[4](https://research.atmos.ucla.edu/tcd/ssa/guide/mann/mann4.html) Bandwidth selection remains a weak point: to resolve two peaks spaced approximately \\( 2W_{0} \\) Hz apart, W must be less than \\( W_{0} \\), severe consequences can follow from choosing too large or too small a bandwidth, and no systematic selection method existed until recently; the ubiquitous NW = 4 default traces to Thomson's original short series (N = 100) and is built into canned software estimators.[8](https://www.osti.gov/pages/servlets/purl/1402465) Significant line components also bias the quadratic estimate, which spectral reshaping addresses.[14](https://fanchung.ucsd.edu/ron/papers/07_08_multitaper.pdf)\n\nAgainst **Welch's method**, which divides the series into overlapping tapered sections and averages their periodograms, multitaper provides a more favorable tradeoff between narrow-band bias, broad-band bias, and variance.[3](https://www.embs.org/tbme/articles/review-multitaper-spectral-analysis/) Welch's sections limit frequency resolution to the section length, whereas multitaper applies orthogonal DPSS tapers to the entire series, giving higher resolution at the same bias and variance.[13](https://arxiv.org/html/2209.15027) Some authors nevertheless prefer a multitaper-Welch hybrid to control variance when larger bandwidths and sample sizes are available.[8](https://www.osti.gov/pages/servlets/purl/1402465) The p = 1, K = 1 case reduces to the Blackman-Tukey case of a single tapered DFT.[4](https://research.atmos.ucla.edu/tcd/ssa/guide/mann/mann4.html)\n\n## References\n\n1. [Thomson's Multitaper Method Revisited](https://arxiv.org/html/2103.11586v1)\n2. [D.J. Thomson (1982). Spectrum estimation and harmonic analysis. Proceedings of the IEEE.](https://doi.org/10.1109/proc.1982.12433)\n3. [A Review of Multitaper Spectral Analysis (Babadi & Brown, IEEE Transactions on Biomedical Engineering, 2014)](https://www.embs.org/tbme/articles/review-multitaper-spectral-analysis/)\n4. [SSA-MTM Toolkit User's Guide: MTM Theory](https://research.atmos.ucla.edu/tcd/ssa/guide/mann/mann4.html)\n5. [spectrum.mtm, spectrum 0.10.0 documentation](https://pyspectrum.readthedocs.io/en/latest/_modules/spectrum/mtm.html)\n6. [Improving Harmonic Analysis Using Multitapering: Precise Frequency Estimation of Stellar Oscillations Using the Harmonic F-test (The Astronomical Journal, 2025)](https://iopscience.iop.org/article/10.3847/1538-3881/adc9b4/meta)\n7. [multitaper: Spectral Analysis Tools using the Multitaper Method (R package documentation)](https://cran.r-project.org/web/packages/multitaper/multitaper.pdf)\n8. [Optimal Bandwidth for Multitaper Spectrum](https://www.osti.gov/pages/servlets/purl/1402465)\n9. [Lees and Park, multitaper in geophysics (Computers & Geosciences)](https://www.leesj.sites.oasis.unc.edu/FETCH/GRAB/Papers/leesandParkMTM.pdf)\n10. [M2NuFFT, A computationally efficient suboptimal power spectrum estimator for fast exploration of nonuniformly sampled time series (Digital Signal Processing, 2025)](https://www.sciencedirect.com/science/article/abs/pii/S1051200425008504)\n11. [Multitaper Spectral Analysis (BIRS workshop report 22w2230)](https://www.birs.ca/workshops/2022/22w2230/report22w2230.pdf)\n12. [Spectrum Estimation and Harmonic Analysis (Thomson, 1982, full text PDF)](https://www.math.ucdavis.edu/~saito/data/ONR15/thomson_spect-est-harm-anal.pdf)\n13. [Improving Power Spectrum Estimation using Multitapering: Efficient asteroseismic analyses for understanding stars, the Milky Way, and beyond](https://arxiv.org/html/2209.15027)\n14. [Reducing the bias of multitaper spectrum estimates](https://fanchung.ucsd.edu/ron/papers/07_08_multitaper.pdf)\n15. [Multitaper Spectral Estimation of Power Law Processes (Percival and Walden)](http://staff.washington.edu/dbp/PDFFILES/mseplp.pdf)\n16. [Enhanced Adaptive Sine Multi-Taper Power Spectral Density Estimation for System Performance Evaluation in Low-Frequency Gravitational Wave Detection (Applied Sciences, 2025)](https://www.mdpi.com/2076-3417/15/7/3919)\n17. [Patil et al. 2024, The Astronomical Journal (mtNUFFT)](https://iopscience.iop.org/article/10.3847/1538-3881/ad7029/pdf)\n18. [Park et al. 1987 (JGR), multitaper eigentaper paper](http://www.jmlilly.net/papers/park87a-jgr.pdf)\n19. [Some advances in non-parametric multiple time series and spectral analysis (Environmetrics)](https://onlinelibrary.wiley.com/doi/10.1002/env.3170050308)\n20. [pmc.ncbi.nlm.nih.gov](https://pmc.ncbi.nlm.nih.gov/articles/PMC4179109/)\n\n---\n*Topic: Encyclopedia › Physical world and mathematics › Mathematics and statistics › Statistics and probability › Statistical inference, estimation, sampling, and testing › Estimation theory and estimator families*\n\n*Initially written Sep 29, 2026 · Reviewed: Sep 30, 2026 · Edited: Sep 30, 2026; Oct 21, 2026 · Last review: Sep 30, 2026*\n\n*Copyright 2026 EdgeChat AI, a subsidiary of Biostate AI.*\n\nLicense: Edgepedia Community License 1.0, https://www.edgepedia.org/edgepedia/license\n", "same_as": [ "https://www.math.ucdavis.edu/~saito/data/ONR15/thomson_spect-est-harm-anal.pdf" ], "url": "https://www.edgepedia.org/multitaper", "markdown_url": "https://www.edgepedia.org/multitaper.md", "license": { "name": "Edgepedia Community License 1.0", "url": "https://www.edgepedia.org/edgepedia/license", "summary": "Free with credit, commercial use included. AI training is open to everyone. For other uses, organizations over USD 100M in revenue or 100M monthly users license separately.", "spdx": "LicenseRef-Edgepedia-Community-1.0" }, "credit": "\"Multitaper\", Edgepedia (EdgeChat), https://www.edgepedia.org/multitaper. Edgepedia Community License 1.0.", "credit_md": "\"[Multitaper](https://www.edgepedia.org/multitaper)\", Edgepedia (EdgeChat), [https://www.edgepedia.org/multitaper](https://www.edgepedia.org/multitaper). [Edgepedia Community License 1.0](https://www.edgepedia.org/edgepedia/license).", "credit_html": "\"Multitaper\", Edgepedia (EdgeChat), https://www.edgepedia.org/multitaper. Edgepedia Community License 1.0.", "speakable": "The multitaper method is a nonparametric spectral estimation technique that reduces variance by averaging periodograms from orthogonal Slepian tapers, reported by D. J. Thomson in 1982." }