论文信息 - Nonlinear LP-norm estimation: part I - on the choice of the exponent, p, where the errors are additive

Nonlinear LP-norm estimation: part I - on the choice of the exponent, p, where the errors are additive

In this paper we shall be concerned with the errors in the nonlinear model as opposed to the actual estimation of the parameter vector A simulation study is reported which indicates an empirical relationship between the optimal exponent p and the kurtosis of a given symmetric error distribution. The results are supported by the application of recently established asymptotic properties of linear LP-norm estimators to the nonlinear case.

A. Money | R. Gonin | R. Gonin | A. H. Money

[1] H. Nyquist. The optimal Lp norm estimator in linear regression models , 1983 .

[2] R. E. Odeh,et al. Algorithm AS 70: The Percentage Points of the Normal Distribution , 1974 .

[3] V. A. Sposito,et al. On the efficiency of using the sample kurtosis in selecting optimal lpestimators , 1983 .

[4] M. R. Spiegel. Mathematical handbook of formulas and tables , 1968 .

[5] I. Barrodale,et al. An Improved Algorithm for Discrete $l_1 $ Linear Approximation , 1973 .

[6] P. R. Fisk,et al. Distributions in Statistics: Continuous Multivariate Distributions , 1971 .

[7] W. R. Buckland,et al. Distributions in Statistics: Continuous Multivariate Distributions , 1973 .

[8] R. Jennrich. Asymptotic Properties of Non-Linear Least Squares Estimators , 1969 .

[9] H. Leon Harter,et al. Nonuniqueness of least absolute values regression , 1977 .

[10] A. Money,et al. The linear regression model: Lp norm estimation and the choice of p , 1982 .

[11] G. Alistair Watson,et al. On an Algorithm for Discrete Nonlinear L1 Approximation , 1971, Comput. J..

[12] P-norm estimators in a linear regression model , 1976 .

[13] G. P. Clarke,et al. Moments of the Least Squares Estimators in a Non‐Linear Regression Model , 1980 .