Expansions for the distribution of M-estimates with applications to the Multi-Tone problem

Christopher S. Withers; Saralees Nadarajah

ESAIM: Probability and Statistics (2012)

  • Volume: 15, page 139-167
  • ISSN: 1292-8100

Abstract

top
We give a stochastic expansion for estimates that minimise the arithmetic mean of (typically independent) random functions of a known parameter θ. Examples include least squares estimates, maximum likelihood estimates and more generally M-estimates. This is used to obtain leading cumulant coefficients of needed for the Edgeworth expansions for the distribution and density n1/2 ( of − θ0) to magnitude n−3/2 (or to n−2 for the symmetric case), where θ0 is the true parameter value and n is typically the sample size. Applications are given to least squares estimates for both real and complex models. An alternative approach is given when the linear parameters of the model are nuisance parameters. The methods are illustrated with the problem of estimating the frequencies when the signal consists of the sum of sinusoids of unknown amplitudes.

How to cite

top

Withers, Christopher S., and Nadarajah, Saralees. "Expansions for the distribution of M-estimates with applications to the Multi-Tone problem." ESAIM: Probability and Statistics 15 (2012): 139-167. <http://eudml.org/doc/222456>.

@article{Withers2012,
abstract = { We give a stochastic expansion for estimates $\widehat\{\theta\}$ that minimise the arithmetic mean of (typically independent) random functions of a known parameter θ. Examples include least squares estimates, maximum likelihood estimates and more generally M-estimates. This is used to obtain leading cumulant coefficients of $\widehat\{\theta\}$ needed for the Edgeworth expansions for the distribution and density n1/2 ( of \widehat\{\theta\}$ − θ0) to magnitude n−3/2 (or to n−2 for the symmetric case), where θ0 is the true parameter value and n is typically the sample size. Applications are given to least squares estimates for both real and complex models. An alternative approach is given when the linear parameters of the model are nuisance parameters. The methods are illustrated with the problem of estimating the frequencies when the signal consists of the sum of sinusoids of unknown amplitudes. },
author = {Withers, Christopher S., Nadarajah, Saralees},
journal = {ESAIM: Probability and Statistics},
keywords = {Bias; edgeworth; maximum likelihood; M-estimates; Skewness; bias; Edgeworth; -estimates; skewness},
language = {eng},
month = {1},
pages = {139-167},
publisher = {EDP Sciences},
title = {Expansions for the distribution of M-estimates with applications to the Multi-Tone problem},
url = {http://eudml.org/doc/222456},
volume = {15},
year = {2012},
}

TY - JOUR
AU - Withers, Christopher S.
AU - Nadarajah, Saralees
TI - Expansions for the distribution of M-estimates with applications to the Multi-Tone problem
JO - ESAIM: Probability and Statistics
DA - 2012/1//
PB - EDP Sciences
VL - 15
SP - 139
EP - 167
AB - We give a stochastic expansion for estimates $\widehat{\theta}$ that minimise the arithmetic mean of (typically independent) random functions of a known parameter θ. Examples include least squares estimates, maximum likelihood estimates and more generally M-estimates. This is used to obtain leading cumulant coefficients of $\widehat{\theta}$ needed for the Edgeworth expansions for the distribution and density n1/2 ( of \widehat{\theta}$ − θ0) to magnitude n−3/2 (or to n−2 for the symmetric case), where θ0 is the true parameter value and n is typically the sample size. Applications are given to least squares estimates for both real and complex models. An alternative approach is given when the linear parameters of the model are nuisance parameters. The methods are illustrated with the problem of estimating the frequencies when the signal consists of the sum of sinusoids of unknown amplitudes.
LA - eng
KW - Bias; edgeworth; maximum likelihood; M-estimates; Skewness; bias; Edgeworth; -estimates; skewness
UR - http://eudml.org/doc/222456
ER -

References

top
  1. J.G. Booth, P. Hall and A.T.A. Wood, On the validity of Edgeworth and saddlepoint approximations. Journal of Multivariate Analysis51 (1994) 121–138.  Zbl0807.62015
  2. L. Comtet, Advanced Combinatorics. Reidel, Dordrecht, Holland (1974).  Zbl0283.05001
  3. R. Gatto and E. Ronchetti, General saddlepoint approximations of marginal densities and tail probabilities. Journal of the American Statistical Association91 (1996) 666–673.  Zbl0869.62017
  4. Y. Kakizawa and M. Taniguchi, Higher order asymptotic relation between Edgeworth approximation and saddlepoint approximation. Journal of the Japan Statistical Society24 (1994) 109–119.  Zbl0818.62014
  5. A.C. Monti, A new look at the relationship between Edgeworth expansion and saddlepoint approximation. Statistics and Probability Letters17 (1993) 49–52.  Zbl0765.62025
  6. I.S. Reed, On a moment theorem for complex Gaussian processes. IRE Transactions on information theoryIT-8 (1962) 194–195  Zbl0102.34903
  7. C.S. Withers and S. Nadrajah, The bias and skewness of (univariate) M-estimates in regression. Technical Report, Applied Mathematics Group, Industrial Research Ltd., Lower Hutt, New Zealand (2007).  
  8. C.S. Withers and S. Nadarajah, Tilted Edgeworth expansions for asymptotically normal vectors. Annals of the Institute of Statistical Mathematics, doi: (2008).  Zbl06537882DOI10.1007/s10463-008-0206-0

NotesEmbed ?

top

You must be logged in to post comments.

To embed these notes on your page include the following JavaScript code on your page where you want the notes to appear.

Only the controls for the widget will be shown in your chosen language. Notes will be shown in their authored language.

Tells the widget how many notes to show per page. You can cycle through additional notes using the next and previous controls.

    
                

Note: Best practice suggests putting the JavaScript code just before the closing </body> tag.