Sufficient Conditions for the Consistency of Maximum Likelihood Estimation Despite Misspecification of Distribution in Multinomial Discrete Choice Models

Ruud, Paul A.

doi:10.2307/1912257

Cited by 103 publications

(59 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Formally, this result can be seen as an extension of Wooldridge's (2002) analytical result of for Probit, claiming Lee's (1982) and Ruud's (1983) approximation results for Logit. Formally, Eq.…”

Section: Simulation and Forecasting With The Control-function Methodssupporting

confidence: 60%

“…However if the sample is large enough, it is first possible to claim the Central Limit Theorem to say that v +e will be normally distributed. The argument is completed using the results from Lee (1982) and Ruud (1983), which state that the approximation of a Normal by an Extreme Value distribution causes only negligible discrepancies. The Monte Carlo experiments reported in Section 5 will show that this approximation is also valid in the study of the change of scale with the application of the control-function method.…”

Section: Change Of Scale In the Application Of The Control-function Mmentioning

confidence: 99%

See 1 more Smart Citation

Change of Scale and Forecasting with the Control-Function Method in Logit Models

Guevara

Ben‐Akiva

2012

Transportation Science

View full text Add to dashboard Cite

The control-function method is the most suitable tool to address endogeneity for several discrete choice models that are relevant in transportation research. The estimators obtained with the control-function method are consistent only up to a scale. In this paper we first depict the determinants of this change of scale by adapting an existing result for omitted orthogonal attributes in Logit models. Then, we study the problem of forecasting under these circumstances. We show that a procedure proposed in previous literature may lead to significant biases, and we suggest novel alternatives to be used with synthetic populations. We also discuss potential extensions of these results to other non-Logit discrete choice models. We use Monte Carlo experimentation and real data on residential location choice to demonstrate these results. The paper finishes summarizing the findings of this research and suggesting future lines of research in this area.

show abstract

Section: Simulation and Forecasting With The Control-function Methodssupporting

confidence: 60%

Section: Change Of Scale In the Application Of The Control-function Mmentioning

confidence: 99%

Change of Scale and Forecasting with the Control-Function Method in Logit Models

Guevara

Ben‐Akiva

2012

Transportation Science

View full text Add to dashboard Cite

show abstract

“…Perhaps not surprisingly, as a likelihoodbased estimator, it is also efficient (under the appropriate regularity conditions). The recent econometric literature includes a variety of different approaches to semiparametric estimation in this context, including the papers of Cosslett (1983), Han (1984Han ( , 1988), Horowitz (1992), Ichimura (1986), Manski (1975Manski ( , 1985Manski ( , 1988), Powell, Stock, and Stoker (1989), Ruud (1983Ruud ( , 1986), Sherman (1993), and Stoker (1986). Our estimator is similar to Ichimura's, in that both can be interpreted as minimum distance estimators.…”

Section: A Amentioning

confidence: 99%

An Efficient Semiparametric Estimator for Binary Response Models

Klein¹,

Spady²

1993

Econometrica

640

616

View full text Add to dashboard Cite

“…It is well known (e.g., Goldberger 1980, Ruud 1983, that even if the distribution is misspecified, the coefficients of the propensity score can still be consistently estimated up to an unknown scale under mild conditions. Chung and Goldberger (1984) show that the estimates from least-squares linear regression are proportional to the estimates from the correctly specified probability model under weak 12 distributional assumptions.…”

mentioning

confidence: 99%

Sensitivity of propensity score methods to the specifications

Zhao

2008

Economics Letters

View full text Add to dashboard Cite

Standard-Nutzungsbedingungen:Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden.Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen.Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in Sensitivity of Propensity Score Methods to the Specifications Zhong Zhao D I S C U S S I O N P A P E R S E R I E S ABSTRACT Sensitivity of Propensity Score Methods to the Specifications *Propensity score matching estimators have two advantages. One is that they overcome the curse of dimensionality of covariate matching, and the other is that they are nonparametric. However, the propensity score is usually unknown and needs to be estimated. If we estimate it nonparametrically, we are incurring the curse-of-dimensionality problem we are trying to avoid. If we estimate it parametrically, how sensitive the estimated treatment effects are to the specifications of the propensity score becomes an important question. In this paper, we study this issue. First, we use a Monte Carlo experimental method to investigate the sensitivity issue under the unconfoundedness assumption. We find that the estimates are not sensitive to the specifications. Next, we provide some theoretical justifications, using the insight from Rosenbaum and Rubin (1983) that any score finer than the propensity score is a balancing score. Then, we reconcile our finding with the finding in Smith and Todd (2005) that, if the unconfoundedness assumption fails, the matching results can be sensitive. However, failure of the unconfoundedness assumption will not necessarily result in sensitive estimates. Matching estimators can be speciously robust in the sense that the treatment effects are consistently overestimated or underestimated. Sensitivity checks applied in empirical studies are helpful in eliminating sensitive cases, but in general, it cannot help to solve the fundamental problem that the matching assumptions are inherently untestable. Last, our results suggest that including irrelevant variables in the propensity score will not bias the results, but overspecifying it (e.g., adding unnecessary nonlinear terms) probably will.JEL Classification: C21, C14, C15, C16, C52

show abstract

Sufficient Conditions for the Consistency of Maximum Likelihood Estimation Despite Misspecification of Distribution in Multinomial Discrete Choice Models

Cited by 103 publications

References 5 publications

Change of Scale and Forecasting with the Control-Function Method in Logit Models

Change of Scale and Forecasting with the Control-Function Method in Logit Models

An Efficient Semiparametric Estimator for Binary Response Models

Sensitivity of propensity score methods to the specifications

Contact Info

Product

Resources

About