Richard H. Spady

An efficient semiparametric estimator for binary response models

This paper proposes an estimator for discrete choice models that makes no assumption concerning the functional form of th e choice probability function where this function can be characterized by an index. The estimator is shown to be consistent, asymptotically normally distributed, and to achieve the semiparametric efficiency bound. Monte Carlo evidence indicates that there may be only modest efficiency losses relative to maximum likelihood estimation when the distribution of the disturbances is known and that the small-sample behavior of the estimator in other cases is good. Copyright 1993 by The Econometric Society.

Hedonic Cost Functions for the Regulated Trucking Industry

In industries where physical output varies with respect to attributes or qualities, it is important to take these differences into account in estimating cost functions. This paper presents a hedonic cost function that can be used to take output characteristics into account and applies it to the regulated trucking industry. It is found that failure to take output characteristics into account creates serious specification error and that inferences concerning economies of scale and factor demand differ substantially between the hedonic and nonhedonic formulation of the cost function.

A DERIVED DEMAND FUNCTION FOR FREIGHT TRANSPORTATION

An attempt has been made to improve upon existing specifications and estimations of freight demand by treating transportation as an input in the production process and estimating the derived input demand equations for rail and trucking associated with a general translog cost function. This estimation method also recognizes that rates and shipment characteristics are jointly dependent and takes this inter dependence into account in estimating the demand functions. Two tentative conclusions emerge from the findings of this paper. First, at least with respect to less-than-truckload (LTL), rail and truck transportation are largely independent. Therefore, relatively little modal misallocation of resources between rail and LTL trucking should result from ICC policies that attempt to maintain the value of service pricing structure. Second, the estimated own price elasticities of demand for rail services were sufficiently high to indicate that the railroads might not benefit from blanket rate increases but could benefit instead from selective rate cutting. It is noted that the unique methodology employed and the empirical findings described in this paper need to be verified using different data and different time periods.

Random Reservation Prices and Bidding Behavior in OCS Drainage Auctions

R E C E N T surveys of the economic literature on auctions1 indicate that there has been a great deal of theoretical work on the properties of equilibrium behavior in auctions and on the design of auctions to maximize the sellers revenues in the presence of optimizing bidders. However, there has been very little empirical work, using field data, that attempts to test various behavioral theories or, indeed, to see whether they are of any practical interest. There has been some good experimental work2 but the relatively small stakes involved in these experimental games, and the relative inexperience of the subjects (given the potential complexity of equilibrium strategies), leave a useful role for the careful study of a detailed field data set. This article analyzes such a data set, the first-price, sealed-bid auctions of drainage and development leases on federal land in

The Performance of Empirical Likelihood and its Generalizations

Abstract. We calculate higher-order asymptotic biases and mean squared errors (MSE) for a simple model with a sequence of moment conditions. In this setup, generalized empirical likelihood (GEL) and infeasible optimal GMM (OGMM) have the same higher-order biases, with GEL having an MSE that exceeds OGMMs by an additional term of order (M i1)=N , i.e. the degree of overidenti cation divided by sample size. In contrast, any 2-step GMM estimator has an additional bias relative to OGMM of order (M i 1)=N and an additional MSE of order (Mi1)2=N: Consequently GEL must be expected to dominate 2-step GMM. In our simple model all GELs have equivalent next higher order behavior because generalized third moments of moment conditions are assumed to be zero; we explore, in further analysis and simulations, the implications of dropping this assumption.

Confidence intervals in generalized method of moments models

We consider the construction of confidence intervals for parameters characterized by moment restrictions. In the standard approach to generalized method of moments (GMM) estimation, confidence intervals are based on the normal approximation to the sampling distribution of the parameters. There is often considerable disagreement between the nominal and actual coverage rates of these intervals, especially in cases with a large degree of overidentification. We consider alternative confidence intervals based on empirical likelihood methods which exploit the normal approximation to the Lagrange multipliers calculated as a byproduct in empirical likelihood estimation. In large samples such confidence intervals are identical to the standard GMM ones, but in finite samples their properties can be substantially different. In some of the examples we consider, the proposed confidence intervals have coverage rates much closer to the nominal coverage rates than the corresponding GMM intervals.

Duopoly with Both Ruin and Entry

Duopoly is modeled here as a prisoners dilemma repeated in continuous time. Firms wish to maximize discounted flows of dividend payments, which are paid when capitalization levels permit. A firm is ruined when its capitalization falls below zero, but each ruined firm is immediately replaced by a new entrant. Attempting to ruin a rival is not necessarily an attractive strategy here, since the postentry game against the new rival may be a less favorable one. On the other hand, firms that are close to ruin have little to lose by playing aggressively and, hence, are attractive targets of aggression themselves. Equilibria are constricted that reflect these considerations.

The Catholics and the Others: The Denominational Backdrop to Modern American Politics

Abstract This paper goes in search of the contribution of the five major religious families in American society to the ideological landscape for electoral conflict. Taking advantage of new methodological opportunities and a rich but underutilized dataset, it considers the distribution of political values within denominations, the link between these values and voting behavior, and the strategic landscape – plus strategic dilemmas – that results from that link. By considering these across the most recent quarter-century, it isolates an older world sometimes characterized as “the Protestant nation,” where Catholics offered additionally distinctive political behavior, and a new world in which the great denominations behave very differently, but where changes in Catholic behavior are arguably most critical to this change.