Lung-fei Lee

THE MEASUREMENT AND SOURCES OF TECHNICAL INEFFICIENCY IN THE INDONESIAN WEAVING INDUSTRY

Production function models are estimated with a time series of cross-section data on Indonesian weaving establishments. The sources of technical inefficiency are investigated. Three firm attributes are identified as being potentially related to firm efhciency. Tbey are firm ownership, age and size. The importance of these attributes as sources of inefficiency in the Indonesian weaving industry is investigated and the implications of the find,ings discussed.

Some Approaches to the Correction of Selectivity Bias

This article addresses the issue of specification of econometric selectivity models and suggests approaches for the correction of selectivity bias. Our approaches provide ways to specify selectivity models without the assumption of multinormal distribution. Some flexible function forms for the correction of selectivity bias in the regression equation are derived. All the models considered can be estimated by simple consistent two stage methods. Our approaches provide simple procedures for the testing of selectivity bias without imposing restrictive distributional assumptions and also tests for the normality assumption.

Estimation of some limited dependent variable models with application to housing demand

Abstract A model which extends the switching regression models and combines several different limited dependent variable models into a general framework is introduced. Methods to get consistent estimates and asymptotic efficient estimates are derived. Our estimation procedures are then used to study a housing expenditure model which takes into account the simultaneous determination of whether or not to own, and how much to spend.

Asymptotic Covariance Matrices of Two-Stage Probit and Two-Stage Tobit Methods for Simultaneous Equations Models with Selectivity

The paper discusses the two-stage estimation method for switching simultaneous equations models where the criterion function determining the switching is of the probit type and the tobit type. It derives the asymptotic covariance matrices of these estimators and shows that when the criterion function is of the probit type the correct covariance matrix is underestimated when the heteroscedasticity introduced in the first step is ignored, whereas the same is not necessarily the case for one of the regimes when the criterion function is of the tobit type.

Best Spatial Two‐Stage Least Squares Estimators for a Spatial Autoregressive Model with Autoregressive Disturbances

Abstract Estimation of a cross‐sectional spatial model containing both a spatial lag of the dependent variable and spatially autoregressive disturbances are considered. [Kelejian and Prucha (1998)]described a generalized two‐stage least squares procedure for estimating such a spatial model. Their estimator is, however, not asymptotically optimal. We propose best spatial 2SLS estimators that are asymptotically optimal instrumental variable (IV) estimators. An associated goodness‐of‐fit (or over identification) test is available. We suggest computationally simple and tractable numerical procedures for constructing the optimal instruments.

Consistency And Efficiency Of Least Squares Estimation For Mixed Regressive, Spatial Autoregressive Models

Least squares estimation has casually been dismissed as an inconsistent estimation method for mixed regressive, spatial autoregressive models with or without spatial correlated disturbances. Although this statement is correct for a wide class of models, we show that, in economic spatial environments where each unit can be influenced aggregately by a significant portion of units in the population, least squares estimators can be consistent. Indeed, they can even be asymptotically efficient relative to some other estimators. Their computations are easier than alternative instrumental variables and maximum likelihood approaches.

The effects of improved nutrition, sanitation, and water quality on child health in high-mortality populations☆

Abstract A framework is set out for estimating the effects of interventions on child health that considers changes in the allocation of family resources, who among children survive (survival selectivity), and changes in the health of surviving children net of family resources. Estimates based on structural-equations semi-parametric models applied to data describing households from rural areas of two low-income countries indicate that conventional reduced-form estimates understate the effectiveness of improving sanitation facilities. This is due to the reduced allocation of household resources to children in households with better facilities but not to mortality selection, which is negligible.

SWITCHING REGRESSION MODELS WITH IMPERFECT SAMPLE SEPARATION INFORMATION - WITH AN APPLICATION ON CARTEL STABILITY

An exogenous switching regression model with imperfect reginme classificationi information is specified and applied to a study of cartel stability. An efficient estimation method is proposed which takes this imperfect information into account. The consequences of misclassification are analyzed. The direction of the least squares bias is derived. An optimal regime classification rule is obtained and compared theoretically and empirically with other classification rules. We then examine the Joint Executive Committee, a railroad cartel in the 1880s. The econometric evidence indicates that reversions to noncooperative behavior did occur for the firms in our sample, and these reversions involve a significant decrease in market price. THIS ARTICLE IS CONCERNED WITH the possibility of estimating a switching regression model and its application to a study on cartel stability. The switching regression model is the exogenous switching model proposed by Quandt [24], which generalized a problem of mixture distributions (Day [4]). The sample in this model is generated from distinct regression equations or regimes for each time period. If the investigator has a priori information on how the sample is partitioned into the underlying regimes, it is a switching regression model with known sample separation; otherwise, it is a model with unknown sample separation. Estimation of these latter models has been considered by Quandt [24]. Goldfeld and Quandt [8], and Kiefer [14, 16], among others. A switching regression model is appropriate for the study of cartel behavior when there are price wars, as firms will revert from cooperative to noncooperative behavior, and so the industry supply function shifts occasionally. This model will allow us to exploit the fact that there will be periodic stochastic switches or reversions between cooperative and noncooperative conduct, in order to identify collusive episodes.

Specification and estimation of social interaction models with network structures

This paper considers the specification and estimation of social interaction models with network structures and the presence of endogenous, contextual and correlated effects. With macro group settings, group-specific fixed effects are also incorporated in the model. The network structure provides information on the identification of the various interaction effects. We propose a quasi-maximum likelihood approach for the estimation of the model. We derive the asymptotic distribution of the proposed estimator, and provide Monte Carlo evidence on its small sample performance. Copyright The Author(s). Journal compilation Royal Economic Society 2010.

A SPATIAL DYNAMIC PANEL DATA MODEL WITH BOTH TIME AND INDIVIDUAL FIXED EFFECTS

This paper establishes asymptotic properties of quasi-maximum likelihood estimators for spatial dynamic panel data with both time and individual fixed effects when the number of individuals n and the number of time periods T can be large. We propose a data transformation approach to eliminate the time effects. When n / T → 0, the estimators are null consistent and asymptotically centered normal; when n is asymptotically proportional to T , they are null consistent and asymptotically normal, but the limit distribution is not centered around 0; when n / T → ∞, the estimators are consistent with rate T and have a degenerate limit distribution. We also propose a bias correction for our estimators. When n 1/3 / T → 0, the correction will asymptotically eliminate the bias and yield a centered confidence interval. The estimates from the transformation approach can be consistent when n is a fixed finite number.