# A third order optimum property of the ML estimator in a linear functional relationship model and simultaneous equation system in econometrics

Article

First Online:

## Summary

The maximum likelihood (ML) estimator and its modification in the linear functional relationship model with incidental parameters are shown to be third-order asymptotically efficient among a class of almost median-unbiased and almost mean-unbiased estimators, respectively, in the large sample sense. This means that the limited information maximum likelihood (LIML) estimator in the simultaneous equation system is third-order asymptotically efficient when the number of excluded exogenous variables in a particular structural equation is growing along with the sample size. It implies that the LIML estimator has an optimum property when the system of structural equations is large.

## Key words and phrases

Maximum likelihood estimator third-order efficiency linear functional relationship incidental parameter LIML estimator large econometric models## Preview

Unable to display preview. Download preview PDF.

## References

- [1]Akahira, M. and Takeuchi, K. (1981).
*The Concept of Asymptotic Efficiency and Higher Order Efficiency in Statistical Estimation Theory*, No. 7, Springer-Verlag.Google Scholar - [2]Anderson, T. W. (1974). An asymptotic expansion of the limited information maximum likelihood estimate of a coefficient in a simultaneous equation system,
*J. Amer. Statist. Ass.*,**69**, 565–572.MathSciNetCrossRefGoogle Scholar - [3]Anderson, T. W. (1976). Estimation of linear functional relationships: approximate distributions and connections with simultaneous equations in econometrics,
*J. R. Statist. Soc.*,**B38**, 1–38.MathSciNetzbMATHGoogle Scholar - [4]Anderson, T. W. (1984). Estimating linear functional relationships,
*Ann. Statist.*,**12**, 1–45.MathSciNetCrossRefGoogle Scholar - [5]Anderson, T. W. and Sawa, T. (1979). Evaluation of the distribution function of the two-stage least squares estimate,
*Econometrica*,**47**, 163–183.CrossRefGoogle Scholar - [6]Anderson, T. W., Kunitomo, N. and Sawa, T. (1982). Evaluation of the distribution function of the limited-information maximum likelihood estimator,
*Econometrica*,**50**, 1009–1027.MathSciNetCrossRefGoogle Scholar - [7]Anderson, T. W., Kunitomo, N. and Morimune, K. (1986). Comparing single equation estimators in a simultaneous equation system,
*Econom. Theory*,**2**, 1–32.CrossRefGoogle Scholar - [8]Fujikoshi, Y., Morimune, K., Kunitomo, N. and Taniguchi, M. (1982). Asymptotic expansions of the distributions of the estimates of coefficients in a simultaneous equation system,
*J. Econometrics*,**18**, 191–205.MathSciNetCrossRefGoogle Scholar - [9]Fuller, W. A. (1977). Some properties of a modification of the limited information estimator,
*Econometrica*,**45**, 939–953.MathSciNetCrossRefGoogle Scholar - [10]Ghosh, J. K., Sinha, B. K. and Wieand, H. S. (1980). Second order efficiency of the MLE with respect to any bounded bowl-shaped loss function,
*Ann. Statist.*,**8**, 506–521.MathSciNetCrossRefGoogle Scholar - [11]Kunitomo, N. (1980). Asymptotic expansions of the distributions of estimators in a linear functional relationship and simultaneous equations,
*J. Amer. Statist. Ass.*,**75**, 693–700.MathSciNetCrossRefGoogle Scholar - [12]Kunitomo, N. (1981). Asymptotic optimality of the limited information maximum likelihood estimator in large econometric models,
*Economic Studies Quarterly*,**XXXII**, 247–266.Google Scholar - [13]Kunitomo, N. (1982). Asymptotic efficiency and higher order efficiency of the limited information maximum likelihood estimator in large econometric models,
*Tech. Rep. No. 365*, Institute for Mathematical Studies in the Social Sciences, Stanford University.Google Scholar - [14]Kunitomo, N. (1986). Comparing some modified maximum likelihood estimator of a slope coefficient in a linear functional relationship.
*J. Japan Statist. Soc.*,**16**, 173–185.MathSciNetzbMATHGoogle Scholar - [15]Morimune, K. (1983). Approximate distributions of
*k*-class estimators when the degree of overidentifiability is large compared with the sample size,*Econometrica*,**51**, 821–841.MathSciNetCrossRefGoogle Scholar - [16]Morimune, K. and Kunitomo, N. (1980). Improving the maximum likelihood estimate in linear functional relationships for alternative parameter sequences,
*J. Amer. Statist. Ass.*,**73**, 867–871.MathSciNetCrossRefGoogle Scholar - [17]Pfanzagl, J. and Wefelmeyer, W. (1978). A third order optimum property of the maximum likelihood estimator,
*J. Multivariate Anal.*,**8**, 1–29.MathSciNetCrossRefGoogle Scholar - [18]Takeuchi, K. (1972).
*Contributions to the Theory of Statistical Inference in Econometrics*(in Japanese), Toyokeizai-Shinposha, Tokyo.Google Scholar - [19]Takeuchi, K. and Morimune, K. (1985). Asymptotic completeness of the extended maximum likelihood estimators in a simultaneous equation system,
*Econometrica*,**53**, 177–200.MathSciNetCrossRefGoogle Scholar

## Copyright information

© The Institute of Statistical Mathematics, Tokyo 1987