Limited Dependent Variable Research Articles

Estimation of a Taiwan monetary reaction function with time varying parameters

This paper employs a nonlinear Kalman filter to examine the time-varying responses of Taiwan's monetary policy in the presence of a limited dependent variable. The Kalman filtered parameters reveal that the responses are not constant but change over time. Furthermore, a counter-cyclical reaction function is identified and a stronger than usual discretionary policy during the recessions is found.

The Interpretation of Coefficients in N‐Chotomous Qualitative Response Models*

AbstractResearchers in financial accounting often use qualitative response models in choice‐based empirical research. Most of this research relies on the familiar techniques of dichotomous probit or logistic regression. Only a limited amount of this research uses n‐chotomous qualitative response models such as ordered probit or multinomial logistic regression. A potential explanation for this limited use is that the interpretation of model coefficients in qualitative response models with limited dependent variables (dichotomous or n‐chotomous) differs substantially from OLS regression, and econometric texts do not provide a systematic approach to coefficient interpretation. This paper discusses several approaches to interpreting coefficients in n‐chotomous qualitative response models. These methods focus on partial derivatives, elasticities of probability, sensitivity analysis, and odds ratios. The methods are applied to the models presented in Thomas (1989) and Mittelstaedt (1989). Additional analyses of the models demonstrate that the methods of interpretation can provide different conclusions or strengthen existing conclusions. The methods provide a better understanding of the directional effects of model coefficients, the relative responsiveness of the probability of choice to changes in the independent variables, and the effects of changes in the independent variables on the probability of choice. These methods should make these models more attractive to researchers interested in choice‐based financial accounting research, and allow for a broader range of decision outcomes than that provided by dichotomous qualitative response models.

Two‐Step Estimation of a Censored System of Equations

AbstractA consistent two‐step estimation procedure is proposed for a system of equations with limited dependent variables. Monte Carlo simulation results suggest the procedure outperforms an existing two‐step method.

Bayesian inference in a simultaneous equation model with limited dependent variables

This paper develops a Bayesian approach for inference in a simultaneous equation model with limited dependent variables (SLDV). By employing a combination of Gibbs sampling and data augmentation, we can draw from the exact posterior of this SLDV model and avoid direct evaluation of the non-trivial likelihood function. A by-product from our posterior simulation is the Savage—Dickey density ratio which is used for computing the Bayes factor. The practicality and efficiency of the proposed method are illustrated through an example in corporate finance.

The Method of Simulated Scores for the Estimation of LDV Models

The method of simulated scores (MSS) is presented for estimating limited dependent variables models (LDV) with flexible correlation structure in the unobservables. We propose simulators that are continuous in the unknown parameter vectors, and hence standard optimization methods can be used to compute the MSS estimators that employ these simulators. The first continuous method relies on a recursive conditioning of the multivariate normal density through a Cholesky triangularization of its variance-covariance matrix. The second method combines results about the conditionals of the multivariate normal distribution with Gibbs resampling techniques. We establish consistency and asymptotic normality of the MSS estimators and derive suitable rates at which the number of simulations must rise if biased simulators are used.

Monte Carlo evaluation of multivariate normal probabilities

This paper extends research on the simulation of multivariate normal probabilities of high-order dimension by developing a new family of simulators. Such simulators are useful for models with limited dependent variables, including multinomial probit, in panel studies, spatial analysis, and time series analysis. The simulators are derived from a Cholesky decomposition of the covariance matrix, combined with a suitable choice of an importance sampling distribution. The paper studies, among others, the impact of antithetical sampling. The insights gained in this paper are of use in Bayesian analysis as well, in the evaluation of posterior densities.

An R2criterion based on optimal predictors

: The predictor that minimizes mean-squared prediction error is used to derive a goodness-of-fit measure that offers an asymptotically valid model selection criterion for a wide variety of regression models. In particular, a new goodness-of-fit criterion (cr2) is proposed for censored or otherwise limited dependent variables. The new goodness-of-fit measure is then applied to the analysis of duration.

Semi‐parametric estimation of simultaneous equations with limited dependent variables: A case study of female labour supply

AbstractWe propose a general strategy to estimate semi‐parametrically simultaneous equations with limited dependent variables. First, each reduced form (RF) is estimated with various semi‐parametric methods. Second, the specification of each RF is tested to select an appropriate method. Third, the structural form (SF) equations are estimated using minimum distance methods and the restrictions among the SF and RF parameters. A case study of female labour supply is presented.

Binary Outcomes and Endogenous Explanatory Variables: Tests and Solutions with an Application to the Demand for Contraceptive Use in Tunisia

Many demographic studies examine discrete outcomes, and researchers often suspect that some of the explanatory variables may be influenced by the same unobserved factors that determine the discrete outcome under examination. In linear models, the standard solution to this potential endogeneity bias is an estimator such as two-stage least squares. These methods have been extended to models with limited dependent variables, but there is little information on the performance of the methods in the types of data sets typically used in demographic research. This paper helps to fill this gap. It describes a simple analytic framework for estimating the effects of explanatory variables on discrete outcomes, which controls for the potential endogeneity of explanatory variables. It also discusses tests for exogeneity and joint determination of the outcomes and the explanatory variables. It summarizes the results of a Monte Carlo study of the performance of these techniques and uses these results to suggest how researchers should approach these problems in practice. We apply these methods to the examination of the impact of fertility intentions on contraceptive use, based on data from the 1988 Tunisia Demographic and Health Survey.

An econometric analysis of teacher mobility

Teacher migration is analyzed using Missouri beginning teacher survey data. A simultaneous equations model involving qualitative and limited dependent variables is developed. Earnings equations which account for self-selectivity are estimated for migrant and non-migrant teachers. They support the conventional human capital theory of migration. Teacher migration is most strongly influenced by home ownership and the leadership style of the school administrator.

Endogenous Switching Regression Models with Limited Dependent Variables

Social research often involves estimating the effects of a categorical treatment on a dependent outcome variable. Endogenous switching regression models are natural extensions of classical experimental designs, which allow tests of assumptions about the exogeneity of treatment effects from survey data. Endogenous switching regression applications are pervasive in the econometric literature and are becoming widely used by sociologists. However, sociologists have devoted somewhat less attention to endogenous switching models involving limited dependent variables. Switching regression models for continuous variables can be generalized to account for binary and censored dependent variables. These models are applied to estimate the effect of family structure on early family formation.

Individual behaviour and tobacco consumption: a panel data approach.

This paper considers estimators of tobacco demand equations using Becker and Murphy's model of addiction with a complete panel of households for Spain. With these tools, we face two main problems: first, the endogeneity of past and future consumption and, second, the limited-dependent variable. To control these problems simultaneously is difficult and we proceed to confront them separately. We follow an instrumental variable approach (which also allows us to control for measurement errors in variables and non-independent effects) to tackle the first and we use a consistent within-group procedure to obtain the parameter vector of the structural form, once we have estimated T-Tobit models for the reduced form in order to deal with the limited-dependent variable problem.

Bayesian inference in limited dependent variable models. An application to measuring strike duration

AbstractThis paper presents a straightforward set of Bayesian techniques for analyzing models involving limited dependent variables; the techniques are demonstrated in an analysis of Kennan's (1985) data on contract strikes in US manufacturing. The data are analyzed by deriving posterior distributions—including probability distributions—of hazard functions for strike duration using numerical Monte Carlo methods. The distributions are employed to derive coverage intervals for hazard functions, to assess the relative plausibility of nonnested hypotheses concerning the shape of the functions, and to assess the impact of industrial production on duration.

Nonwage benefits as a limited-dependent variable: Implications for the impact of unions

To better delineate the role of unionism in influencing the provision of nonwage benefits, this article argues that benefits should be viewed as a limited-dependent variable. Employing such an approach, the study finds that previous OLS estimates of the union effect embody substantial biases, which vary by industrial and employee-group classification. Moreover, the role of unionism — plan initiation versus ability to augment benefit levels at established plans — varies according to employee group, industrial classification, and the type of benefits.

Amemiya‘s generalized least squares and tests of overidentification in simultaneous equation models with qualitative or limited dependent variables

Amemiya's generalized least squares method for the estimation of simultaneous equation modeis with qualitative or limited dependent variables is known to be efficient relative to many popular two stage estimators. This note points out that test statistics for overidentification restrictions can be obtained as by-products of Amerniya's generalized least squares procedure. Amemiya's procedure is shown to be a minimum chisquare method. The Amemiya procedure is valuable both for efficient estimation and for model evaluation of such models.

Residuals for limited dependent variable models

General residual expressions based on likelihood equations are given for regression models with censoring or truncation on the endogenous variable. Using nonparametric density estimation such general expressions may be used to obtain residuals for models estimated by, e.g., semiparametric estimators. The suggested approach is illustrated by numerical examples.

The economic effects of faculty unions

The relationship between faculty salaries and unionism at Ph.D.-level universities is investigated. Unlike previous attempts at analyzing this topic, the model endogenizes the union-nonunion choice using appropriate techniques for analyzing limited dependent variables. These techniques allow computation of average salaries for all schools under the assumption that they are unionized and also under the assumption that they are not unionized. Subtraction of the predicted nonunion salary from the union salary provides an estimate of salary differences, which is corrected for selectivity bias.

Interviewed by Nicholas M. Kiefer

Econometrics as practiced by Arthur (Art) Goldberger demonstrates extraordinary sensitivity to issues of measurement and model specification, and unusual care and caution in interpretation of results, as well as a thorough and comprehensive mastery of econometric theory. His landmark 1964 book, Econometric Theory, set a new standard of rigor in econometrics, and at the same time treated the important problems posed by limited and qualitative dependent variables years before any other text. Art Goldberger's work ranges from early contributions to macro modeling through demand analysis, multivariate modeling with latent variables, and models for sample selectivity, to his highly regarded work on important social issues of heritability of IQ, effectiveness of public versus private schools, and measurement of salary discrimination. Goldberger's influential work, especially on modeling latent or unobservable variables, is widely known and applied in sociology and psychology as well as in economics. Art has been at Wisconsin for many years, and this association is an important reason for Wisconsin's continuing reputation as a leading center for quantitative social sciences.The quality and influence of Art Goldberger's work has earned him many professional honors. He is a Fellow of the Econometric Society, the American Statistical Association, the American Academy of Arts and Sciences, and the American Association for the Advancement of Science, and has twice been a Guggenheim Fellow. He is a Distinguished Fellow of the American Economic Association and a member of the National Academy of Sciences. Art gave the Woytinsky Lecture at the University of Michigan in 1985.This interview took place on May 5 in Art Goldberger's office overlooking Lake Mendota. Art's remarks cover a wide range of topics, and I hope they are of interest to social scientists generally as well as to econometricians.

Hearing Beginning Readers Read: problems of relating practice to theory in interpretation and evaluation

AbstractResearch reports of the involvement of parents in hearing children read show equivocal results in terms of improvement in reading scores or progress through school books. Analysing the practices of both parents and teachers as reported in the literature this paper demonstrates that the phenomenon is not a simple, fixed treatment meriting assessment on limited dependent variables. It shows that in current evaluations there is insufficient analysis of the social and intentional nature of the practice, and of associated educational and theoretical rationales, to do justice to possible effects and to help interpret various findings.

An application of LDV models to household expenditure analysis in Mexico

In this paper Mexican household survey data is used to estimate the Extended Linear Expenditure System. The estimation is carried out taking into account the non-negative nature of expenditures through the use of Limited Dependent Variable (LDV) models. Substantial differences are found in estimated demand responses between OLS and MLE on the LDV models. If the latter estimation procedure is taken as appropriate the findings show that commonly used OLS results may significantly overestimate the marginal propensity to consume and may considerably underestimate subsistence expenditures.