The Small Sample Properties of Simultaneous Equation Least Absolute Estimators vis-a-vis Least Squares Estimators

Glahe, Fred R.; Hunt, Jerry G.

doi:10.2307/1912204

Cited by 31 publications

(6 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When LS is used in the first stage and LAD in the second stage, this estimator was denoted, a bit misleadingly, two stage least absolute deviations (2SLAD). This case had also been considered much earlier, in a Monte Carlo study by Glahe and Hunt (1970). Their experiment contrasted the 2SLS to the 2SLAD, as well as OLS to LAD estimates in a simultaneous equation model, with a very small sample size (10 and 20 observa tions), and Gaussian (Normal) errors.…”

Section: Introductionmentioning

confidence: 91%

“…where Y l , Y2 are jointly determined endogenous variables; Xi, i = 1"", 4 are exogenous variables, and U l , U 2 are correlated er ror terms, with covariance 0"1 2 > O. This model, or a small variant of it, using a different normalization, was used in Nagar (1960), Mad dala (1974) and others, for 28L8 studies, and Glahe and Hunt (1970), also considering 2SLAD estimation.…”

Section: Description Of Experiments Conductedmentioning

confidence: 99%

“…Both the small sample bias and mean squared error, as well as hypothesis testing performance is investigated. The experiments are based on the set up of Glahe and Hunt (1970), and also Maddala (1974), who investigated the small sample properties of significance tests for the 28L8. The experiments include a variety of error distributions (Gaussian, Lognormal, t, and Cauchy) and misspecification of the empirical model.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Small Sample Evidence of Regression Quantile Estimates for Structural Models: Estimation and Testing

Ribeiro¹

1998

Brazilian Review of Econometrics

View full text Add to dashboard Cite

A small Monte Carlo study is conducted to investigate the small sample properties of regression quantiles estimates in structural econometric models with iid errors. Two versions of median regression (the 50 th quantile) estimates using instrumented regressors are used, with the fi rst stage estimated by least squares and by median regression. These estimators are compared to two stage least squares, ordinary least squares and simple median regression. It is found that the performance of the two stage regression quantile estimators is equal, if not better, than two stage least squares in terms of bias and that t tests have good size and power properties, under a variety of error distributions. Resumo Atraves de urn pequeno experimento de Monte Carlo, estudamos as pro priedades em pequenas amostras dos estimadores de regressao quantilica para uma equagao estrutural em urn modele de equag5es simultaneas com erros iid. Duas versoes de estimadores de regressao mediana (0 50Q quantil!percentil) us ando variaveis instrumetais sao empregados, estimando 0 primeiro estagio com minimos quadrados ou com regressao mediana. Comparando-os com estimadores de minimos quadrados em dois estagios(MQ2E) e ordinarios, e regressao mediana .. A previous version of this work was part of my dissertation The Effect of Personal Income Taxes on Labor Supply in Brazil: an application of quantile regression at the University of Illinois at Urbana�Champaign and was presented at the XVIII Encontro Brasileiro de Econometria (EBE) at Aguas de Lind6ia, SP 1 Brazil. I would like to thank the EBE session participants and the refer ees for comments and suggestions and Roger Koenker for very helpful comments. AJiy errors are my own. The financial support of CAPES�Brazil and UFRR at earlier stages of this work is ac knowledged. This paper was supported by CNPq-Brazil under project 301170/96�7. Joaquim Dias and Suzel L. Jensen provided research assistance.

show abstract

Section: Introductionmentioning

confidence: 91%

Section: Description Of Experiments Conductedmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Small Sample Evidence of Regression Quantile Estimates for Structural Models: Estimation and Testing

Ribeiro¹

1998

Brazilian Review of Econometrics

View full text Add to dashboard Cite

show abstract

“…For more on the efficiency of LAD versus OLS, see Glahe and Hunt (1970), Hunt, Dowling, and Glahe (1974), Pfaffenberger and Dinkel (1978), Rosenberg andCarlson (1977), andMitra (1987). The results reported below in the simulation study and the empirical application are based on LAD.…”

Section: Portfolio Formationmentioning

confidence: 99%

Testing Linear Factor Pricing Models With Large Cross Sections: A Distribution-Free Approach

Gungor

Luger

2013

Journal of Business & Economic Statistics

View full text Add to dashboard Cite

We develop a finite-sample procedure to test the beta-pricing representation of linear factor pricing models that is applicable even if the number of test assets is greater than the length of the time series. Our distribution-free framework leaves open the possibility of unknown forms of non-normalities, heteroskedasticity, time-varying correlations, and even outliers in the asset returns. The power of the proposed test procedure increases as the time-series lengthens and/or the cross-section becomes larger. This stands in sharp contrast to the usual tests that lose power or may not even be computable if the crosssection is too large. Finally, we revisit the CAPM and the Fama-French three factor model. Our results strongly support the mean-variance efficiency of the market portfolio. JEL classification: C12, C14, C33, G11, G12 Bank classification: Econometric and statistical methods; Financial markets RésuméLes auteurs élaborent une procédure permettant de tester, en échantillon fini, la représentation des coefficients bêta donnés par les modèles linéaires d'évaluation factorielle, et ce, même si le nombre des actifs dépasse celui des valeurs de la série chronologique. Leur cadre autorise des formes inconnues de distribution autres que la loi normale ainsi que la présence d'hétéroscédasticité, de structures de corrélation variables dans le temps, voire de rendements aberrants. La puissance de la procédure s'accroît avec l'allongement de la série chronologique et la hausse du nombre des actifs. Cette propriété tranche avec les limites des tests habituels, qui perdent de leur puissance ou peuvent même devenir inexécutables si le nombre des actifs est trop élevé. Pour finir, les auteurs réexaminent le modèle d'évaluation des actifs financiers et le modèle trifactoriel de Fama et French. Leurs résultats indiquent clairement que le portefeuille de marché se situe sur la frontière efficiente dans le plan moyenne-variance.

show abstract

“…Glahe and Hunt (1970), Amemiya (1982), and Powell (1983) investigate the two-stage least absolute deviations (2SLAD) estimators that use the least absolute deviations (LAD) estimator instead of the ordinary least squares (OLS) estimator in one or both stages in the two-stage least squares (2SLS) estimation. These estimators thus consistently estimate the structural parameter values, if and only if the population reduced-form regressions satisfy the structural relationship of interest.…”

Section: Article In Pressmentioning

confidence: 99%

Instrumental variable estimation based on conditional median restriction

Sakata¹

2007

Journal of Econometrics

View full text Add to dashboard Cite

The Small Sample Properties of Simultaneous Equation Least Absolute Estimators vis-a-vis Least Squares Estimators

Cited by 31 publications

References 18 publications

Small Sample Evidence of Regression Quantile Estimates for Structural Models: Estimation and Testing

Small Sample Evidence of Regression Quantile Estimates for Structural Models: Estimation and Testing

Testing Linear Factor Pricing Models With Large Cross Sections: A Distribution-Free Approach

Instrumental variable estimation based on conditional median restriction

Contact Info

Product

Resources

About