Maximum Empirical Likelihood Estimation of the Spatial Lag Autoregressive Model

D. M. Lambert
University of Tennessee, Knoxville, TN, USA

Series: Mathematics Research Developments
BISAC: MAT000000

Clear

$0.00

Volume 10

Issue 1

Volume 2

Volume 3

Special issue: Resilience in breaking the cycle of children’s environmental health disparities
Edited by I Leslie Rubin, Robert J Geller, Abby Mutic, Benjamin A Gitterman, Nathan Mutic, Wayne Garfinkel, Claire D Coles, Kurt Martinuzzi, and Joav Merrick

eBook

Digitally watermarked, DRM-free.
Immediate eBook download after purchase.

Product price
Additional options total:
Order total:

Quantity:

Details

This book reformulates the instrumental variable (IV) approach of estimating of the spatial lag autoregressive (SAR) model in the maximum empirical likelihood (MEL) framework, a nonparametric approach to estimate linear and nonlinear regression models. A Monte Carlo analysis compares the precision of the estimators under different levels of heteroskedasticity, lag autocorrelation, and sample size. Results suggest there are no gains in precision from the MEL estimator, and that the SAR-IV estimator is more precise as heteroskedasticity increases. (Imprint: Nova)

ABSTRACT

1. INTRODUCTION

2. GENERALIZED METHOD OF MOMENT ESTIMATION OF THE SAR(1) MODEL

3. MAXIMUM EMPIRICAL LIKELIHOOD ESTIMATION OF THE SAR(1) PROCESS MODEL

4. MONTE CARLO EXPERIMENTS

5. MONTE CARLO RESULTS

CONCLUSIONS

REFERENCES

INDEX

You have not viewed any product yet.