Estymatory odporne w regresji (przegląd)

• Autorzy: Grzegorz Kowalewski

Warianty tytułu

Robust Estimators in Regression (Survey)

• Języki publikacji: PL

Abstrakty

Omówiono trzy główne klasy estymatorów odpornych w regresji:
1. M-estomatory,
2. L-estymatory,
3. R-estymatory.

EN

A frequently arising problem with application of classical statistical method to testing of economic phenomena's relations is the occurrence of outliers. Their appearance in a set of data can completely ruin the advantages of statistical analysis Of such data. Outliers have to be detected and measures have to be taken to reduce their influence upon the statistical analysis results. It can be done with robust estimators.
The paper present 3 most often used class of robust estimators in regression:
1. M-estimators,
2 L-estimators,
3. R-estimators.
You give a detailed description of individual estimators in framework of each class. (original abstract)

Słowa kluczowe

PL

Analiza regresji; Estymacja; Regresja liniowa

EN

Regression analysis; Estimation; Linear regression

• Czasopismo: Prace Naukowe Akademii Ekonomicznej we Wrocławiu. Informatyka i Ekonometria
• Rocznik: 1996
• Tom: 1
• Numer: nr 718 Zastosowania metod ilościowych
• Strony: 43-53

Twórcy

autor

Grzegorz Kowalewski

Bibliografia

• Andrews D.F.: A robust method for multiple linear regression. "Technometrics" 1974 nr 16 (523-531).
• Beaton A.E., Tukey J.W.: The fiting of power series, meaning polynomials, illustrated on band-spectroscopic data. "Technometrics" 1974 nr 16 (147-185).
• Bickel P.J.: On some analogues to linear combinations of order statistics in the linear model. "The Annals of Statistics" 1973 nr 1 (597-616).
• Claerbout J.F., Muir F.: Robust modelling with erratic data. "Geophysics" 1973 nr 38 (826-844).
• Hampel F.R., Ranchetti E.M., Rousseeuw P.J., Stahl W.A.: Robust Statistis: The Approach Based on Influenced Functions. New York: Wiley & Sons 1986.
• Hoaglin D.C., Mosteller F., Tukey J.W. (red.): Understanding Robust and Exploratory Data Analysis. New York: Wiley & Sons 1983.
• Huber P.J.: Robust regression: asymptotics, conjectures, and Monte Carlo. "The Annals of Statistics" 1973 nr 1 (799-821).
• Jaeckel L.A.: Estimating regression coefficients by minimalizing the dispersion of the residuals. "The Annals of Mathematical Statistics" 1972 nr 43 (1449-1458).
• Johns M.V.: Robust Pitman-like estimators. [w:] R.L. Launer, G.N. Wilkinson (red.) Robustness in Statistics, New York: Academic Press 1979 (49-60).
• Koenker R., Bassett G.: Regression quantiles. "Econometrica" 1978 nr 46 (33-50).
• Lax D.A.: An interim report of a Monte Carlo study of robust estimators of width Technical Report 93, Series 2. Princeton: Princeton University, Department of Statistics 1975.
• Rousseeuw P.J.: Least median of squares regression. "Journal of the American Statistical Association" 1984 nr 79 (871-880).
• Rousseeuw P.J., Leroy A.M.: Robust Regression and Outlier Detection. New York: Wiley & Sons 1987.
• Rousseeuw P.J., Yohai V.: Robust regression by means of S-estimators. [w:] J. Franke, W. Hardie, R.D. Martin (red.): Robust and Nonlinear Time Series Analysis. New York: Springer Verlag 1984 (256-272).
• Ruppert D., Carroll R.J.: THmmed least squares estimations in the linear model. "Journal of the American Statistical Association" 1980 nr 75 (828-838).
• Theil H.: Zasady ekonometrii. Warszawa: PWN 1979.
• Wilcoxon F.: Individual comparisons by ranking methods. "Biometrica" 1945 nr 1 (80-83).
• Wolfowitz J.: The minimum distance method. "The Annals of Mathematical Statistics" 1957 nr 28 (75-88).