Thomas W. Zuehlke

On the choice of functional form for hedonic price functions

The empirical results presented in this paper demonstrate the usefulness of quadriatic models in the estimation of hedonic price functions. We propose a quadriatic semi-log model, a restricted case of the quadriatic Box-Cox model, whose standardized marginal attribute prices are linear functions of the attribute vector. This specification outperforms the linear Box-Cox specification in terms of explanatory power, and captures much of the increase in explanatory power provided by the quadriatic Box-Cox model without the corresponding loss in the ability to interpret the coefficient estimates. For policy applications, the quadriatic Semi-log model provides a useful alternative to linear specification.

Litigation and Settlement: An Empirical Approach

Litigants in civil lawsuits involving monetary damages often find an out-of-court settlement preferable to a trial. Most theoretical models of settlement choice employ an expected-utility-maximizing framework that emphasizes the importance of risk preferences, litigation costs, and the distribution of trial awards. An empirical model of settlement choice is used to examine whether the variables prominent in the theoretical literature are statistically useful in explaining the occurrence and monetary value of settlements. Estimates from a sample of civil filings provide new empirical evidence of how the legal system affects the behavior of litigants during the settlement process. Copyright 1989 by MIT Press.

A Comparison of Two-Stage Estimators of Censored Regression Models

This paper presents a Monte Carlo comparison of the small-sample performance of subsample ordinary least squares, the Heckman-Lee two-stage estimator, and the robust estimator of Lee. Each estimator is considered under bivariate normal, t, and chi-square error structures. The estimates indicate that the Heckman-Lee and Lee estimators do not provide an unequivocal mean square error improvement upon subsample ordinary least squares in small samples. While effectively controlling for selectivity bias, the two-stage estimators suffer a substantial loss of small-sample precision relative to subsample ordinary least squares. Copyright 1991 by MIT Press.

Business Cycle Duration Dependence Reconsidered

Sichel estimated a Weibull hazard model using the National Bureau of Economic Research business cycle chronology and found evidence of duration dependence only for prewar expansions and postwar contractions. The article updates the postwar sample through the end of the most recent expansion and uses a generalized Weibull model that provides much greater flexibility at the expense of one additional parameter. This model finds evidence of duration dependence for all samples and is statistically superior to the conventional Weibull model for all samples except postwar contractions.

The Timing of Out-of-Court Settlements

Parties to a lawsuit will frequently delay litigation, even in circumstances where a voluntary agreement is eventually reached. This article is concerned with what causes legal disputes to be prolonged over time. We discuss dynamic models of litigation to illustrate how changes in the bargaining environment might generate empirical hazard functions. Comparative statics results are then corroborated with empirical estimates of a hazard function adjusted to account for both the heterogeneity of lawsuits and the nonproportional time dependence suggested by theory.

Duration dependence testing for speculative bubbles

Prior research has employed a number of methods to test for speculative bubbles in asset prices, including a method based on the concept of duration dependence. This study explores whether duration dependence tests for speculative bubbles are sensitive to specification decisions. Our results question the efficacy of using measures of duration dependence to test for speculative bubbles. In particular, we find that evidence of duration dependence is sensitive to the method of correcting for discrete observation of continuous duration, the use of value-weighted versus equally weighted portfolios, and the use of monthly versus weekly runs of abnormal returns. (JEL C41, G12)

Estimation of a Tobit model with unknown censoring threshold

Conventional wisdom suggests that only the estimated intercept is affected by imposition of a zero censoring threshold on a Tobit model. This is true for Heckman-Lee estimation. For maximum likelihood (ML) estimation, however, it is only true if the censoring threshold is known and is subtracted from the dependent variable. Failure to properly transform the dependent variable prior to ML estimation of a zero threshold Tobit model will generally bias the coefficient estimates. A long neglected topic is ML estimation of a Tobit model with common, but unknown, censoring threshold. This paper shows that the ML estimator of the censoring threshold is the minimum order statistic from the observed subsample, and that existing software for estimation of a zero-threshold Tobit model is easily adapted to include estimation of the censoring threshold.

Tests of the rational expectations-permanent income hypothesis for developing economies☆

Abstract As part of a larger macroeconomic model of the Peoples Republic of China, Chow (1985) finds no empirical evidence to reject the rational expectations-permanent income hypothesis (REPIH). Chows findings for China challenge conventional wisdom, which questions the validity of the REPIH for developing economies because of limited information flows and the absence of developed credit markets. This paper employs the methodology developed by Hall (1978) and Flavin (1981) in a multicountry study of the REPIH for developing economies. Statistical evidence to reject the REPIH is found for each of the countries considered.

A Search Model of Housing Market Transactions

arrive in a temporal sequence. If an offer is received, a seller must decide whether to accept the current offer, or to reject that offer and continue to search. It is this strategic choice, and the possibility of not receiving an offer, that results in extended selling periods and stocks of unsold houses. Rejection of offers and extended selling periods are to be expected in an uncertain environment, where search may be used productively to obtain a higher selling price. This paper considers the implications of a search environment for measurement of hedonic prices. Because an offer is observed only when it is acceptable to the seller, sold houses constitute a self-selected subsample in a search environment. Accurate measurement of hedonic prices in this situation may be obtained with the two-stage procedure of Heckman [3] and Lee [5], or the maximum likelihood procedure summarized in Amemiya [1]. The empirical model presented in section II is a generalization of the usual hedonic model in the sense that the pricing equation is supplemented with an exchange equation which determines probability of sale. The coefficients in each equation provide useful information regarding housing search. The coefficients in the pricing equation are marginal attribute prices, while the attribute coefficients in the exchange equation are the standardized differences between the marginal attribute prices of buyers and sellers. An interesting implication of this model is that attribute variation will not be systematically related to probability of sale if the marginal attribute prices of buyers and sellers equalize. Thus, this model may be used to address two related questions regarding hedonic analysis in a search environment. First, does competition among buyers and sellers equalize marginal attribute prices in a search environment? And second, is accurate measurement of hedonic prices affected by the probabilistic nature of sale?

Duration Dependence in Real Estate Investment Trusts

Hazard models are used to test for duration dependence in the market for real estate investments trusts. Duration dependence implies an ability to predict the turning points of a cycle. In a sense, these models attempt to predict the timing of mean reversion of the market indices. Since the only sample information used in these tests is the length of time between turning points of the cycle, this methodology avoids the more challenging task of modelling the quantitative values of the series, and should provide relatively robust results because of the relatively weak structure imposed on the estimation process. Empirical evidence of duration dependence is found for all samples except mortgage REIT expansions.