Estimating Nonlinear Dynamic Models Using Least Absolute Error Estimation
- 1 March 1991
- journal article
- research article
- Published by Cambridge University Press (CUP) in Econometric Theory
- Vol. 7 (1) , 46-68
- https://doi.org/10.1017/s0266466600004230
Abstract
We consider least absolute error estimation in a dynamic nonlinear model with neither independent nor identically distributed errors. The estimator is shown to be consistent and asymptotically normal, with asymptotic covariance matrix depending on the errors through the heights of their density functions at their medians (zero). A consistent estimator of the asymptotic covariance matrix of the estimator is given, and the Wald, Lagrange multiplier, and likelihood ratio tests for linear restrictions on the parameters are discussed. A Lagrange multiplier test for heteroscedasticity based upon the absolute residuals is analyzed. This will be useful whenever the heights of the density functions are related to the dispersions.Keywords
This publication has 38 references indexed in Scilit:
- Nonlinear Statistical ModelsWiley Series in Probability and Statistics, 1987
- Least absolute deviations estimation for the censored regression modelJournal of Econometrics, 1984
- A general approach to lagrange multiplier model diagnosticsJournal of Econometrics, 1982
- On convergence of LAD estimates in autoregression with infinite varianceJournal of Multivariate Analysis, 1982
- A note on studentizing a test for heteroscedasticityJournal of Econometrics, 1981
- Trimmed Least Squares Estimation in the Linear ModelJournal of the American Statistical Association, 1980
- Approximation Theorems of Mathematical StatisticsPublished by Wiley ,1980
- Asymptotic Theory of Least Absolute Error RegressionJournal of the American Statistical Association, 1978
- A New Test for HeteroskedasticityJournal of the American Statistical Association, 1969
- Density Estimation for Statistics and Data AnalysisPublished by Springer Nature ,1400