A study of estimator densities and performance under misspecification

Rhodes, George F.; Westbrook, M. Daniel

doi:10.1016/0304-4076(81)90033-6

Cited by 26 publications

(13 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…From Table 1 the following comments can be made. For the TSLS estimator the conclusions are similar to those of Rhodes and Westbrook (1981):…”

Section: Properties O F the Misspecified Distributionssupporting

confidence: 75%

“…'E:~+, -+ + I -+ + nll 7c12 and 77: , 1' 77: , + were chosen from wider preliminary analysis, and in particular, clarify and illustrate the effects of misspecification. In general, the parameter values chosen, are similar to those chosen by Rhodes and Westbrook (1981).…”

Section: Properties O F the Misspecified Distributionsmentioning

confidence: 99%

“…Mariano and Ramage (1980) consider other types of misspecification including the omission of relevant endogenous variables and the misclassification of endogenous regressors as exogenous. Rhodes and Westbrook (1981) compute the exact density function of the TSLS estimator, when exogenous variables are wrongly excluded from the equation being estimated and when there is only one endogenous regressor. The misspecified distributions are compared with the correctly specified ones on the basis of density function concentration (that is, the length of 90% prohability intervals), and location around the true parameter value (that is, the midpoint of the probability interval).…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Finite-sample properties of limited-iinformation estimators in misspecified simultaneous equation models

Lye

1991

Journal of Statistical Computation and Simulation

View full text Add to dashboard Cite

The two most common limited-information estimators in Simultaneous Equation Models are the two-stage least squares and limited-information maximum likelihood estimators. As both of these estimators are complicated functions of the underlying random variables, their exact distributions are difficult to derive. Consequently, their use was first justified on the basis of large sample criteria, such as consistency and asymptotic efficiency. However, in the early 1960s the analysis of the exact distributions and moments of these estimators began, and since this time substantial progress has been made. Although these estimators are asymptotically equivalent, recent research has shown that their finite-sample properties are substantially different. However, the majority of this research has simply concentrated on a correctly specified system of equations, even though, since typically in applied studies theory provides some guidance but falls short of specifying the precise form of structural relationship, the possibilities for misspecification in simultaneous equation models are numerous. The objective of this paper is to extend the finite-sample analysis of these two estimators to include various cases of misspecification.

show abstract

“…From Table 1 the following comments can be made. For the TSLS estimator the conclusions are similar to those of Rhodes and Westbrook (1981):…”

Section: Properties O F the Misspecified Distributionssupporting

confidence: 75%

Section: Properties O F the Misspecified Distributionsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Finite-sample properties of limited-iinformation estimators in misspecified simultaneous equation models

Lye

1991

Journal of Statistical Computation and Simulation

View full text Add to dashboard Cite

show abstract

“…Structural misspecification also results in a misspecified mean: see Hale, Mariano, and Ramage [20], Rhodes and Westbrook [33] and Phillips [28,29]. Thus the results on structural misspecification can be utilized to analyze the effects of non-normality.…”

Section: Step 2 Compute the Unconditional Densitymentioning

confidence: 99%

“…This matrix has in the normal case a central Wishart distribution but due to normality and the misspecification of the mean of the j/s it now has a noncentral Wishart distribution with degrees of freedom / = rank (A) and noncentrality parameter matrix given by Due to non-normality, this matrix has full rank (i.e., 2), and thus the results of Richardson [34] and Sawa [36] are inapplicable. Instead we can use the results of Rhodes and Westbrook [33] on structural misspecification since as we have noted the non-normality misspecifies the mean, which is analogous to the misspecification case. Thus from Rhodes and Westbrook [33] or Phillips [28 9 29] we have the pdf of p\rj given by…”

Section: Analysis Of the Leading Case Smentioning

confidence: 99%

Non-Normal Errors and the Distribution of OLS and 2SLS Structural Estimators

Knight

1986

Econ Theory

View full text Add to dashboard Cite

This paper examines the sensitivity of the distributions of OLS and 2SLS estimators to the assumption of normality of disturbances in a structural equation with two included endogenous variables. The approach taken is that of imposing Edgeworth distributed errors on the reduced form equations and deriving the pdf of the estimators via the technique of Davis [11]. The sensitivity of the pdf s to changes in the non-normality parameters, i.e., skewness and kurtosis is examined via extensive numerical computations.The finite-sample properties of econometric estimators have, in recent years, attracted a great deal of attention. (See Phillips [28 9 29] for an excellent overview). Although we now know the exact density function for the instrumental variable estimator in the general n + 1 endogenous variable case and structural misspecification has been extensively analysed, there are very few published results available on the effect of non-normal disturbances on the distribution of econometric estimators, (see [9], [32], and [42]). Knight [23 9 24] has recently analyzed the effect of non-normal disturbances on the moments of OLS and 2SLS estimators, by applying results of

show abstract

Finite-sample properties of single-equation estimators under structural change

Hodoshima

1992

Journal of Econometrics

View full text Add to dashboard Cite

A study of estimator densities and performance under misspecification

Cited by 26 publications

References 10 publications

Finite-sample properties of limited-iinformation estimators in misspecified simultaneous equation models

Finite-sample properties of limited-iinformation estimators in misspecified simultaneous equation models

Non-Normal Errors and the Distribution of OLS and 2SLS Structural Estimators

Finite-sample properties of single-equation estimators under structural change

Contact Info

Product

Resources

About