Econometric analysis often addresses model misspecification due to the improper pooling of observations. One major problem in testing for improper pooling is the requirement that alternative stratifications be obtained from others through sets of restrictions (i.e., that they be nested stratifications), thus eliminating a large class of alternative non-nested stratifications. We propose that non-nested tests can be used to compare non-nested stratifications. We formally define the econometric problem, and show the applicability of the J, JA, Cox and non-nested F tests. We then use the four tests to compare spatial stratifications in a model of a house price determination.

Bilateral trade flows are used to examine the Linder hypothesis and the effect of exchange-rate variability in a gra vity-type trade model derived from an underlying demand and supply mo del. A behavioral model is used to justify examining these issues joi ntly. The model performs well empirically using a sample of seventeen countries for the period 1974-82. The authors find overwhelming supp ort for the Linder hypothesis and this version of the gravity model. Moreover, they find strong support for the hypothesis that increased exchange-rate variability affects bilateral trade flows.

ECONOMISTS and statisticians who construct estimates of total factor productivity or who estimate production functions or systems of consumer demand functions are often forced to aggregate subsets of their data. In order to perform this aggregation, an index number formula is generally used. A price index P(pO, pl, x?, xI) is defined to be a function P of the prices of the N commodities to be aggregated in periods 0 and 1,p?-(pll, . . . , PNO) and pl (pl,.'.. PN'), respectively, and of the corresponding quantities utilized during periods 0 and 1, x? (xi?, . . .,XNO) andX1 _ (xi', . . .,XN1), respectively. A quantity index Q(p0, pl, x?, xl) is defined to be another function Q of the price and quantity vectors for the two periods. Generally, we assume that P and Q satisfy Fisher's (1922) weak factor reversal test:

Gordon C. Rausser, J. W. Freebairn, Estimation of Policy Preference Functions: An Application to U.S. Beef Import Quotas, The Review of Economics and Statistics, Vol. 56, No. 4 (Nov., 1974), pp. 437-449

Exponential regressions are frequently used when outcomes are non-negative. They are attractive because they are easy to interpret and to estimate, using pseudo maximum likelihood (PML). However, the validity of these methods depends on the correct specification of the conditional expectation, and little is known regarding their properties when the conditional expectation is misspecified. We show that PML estimators of misspecified exponential models provide optimal approximations to the conditional expectation, in a weighted mean squared error sense, and we give conditions under which their Poisson PML estimator identifies average marginal effects.

We semiparametrically estimate average causal effects of different lengths of exposure to academic and vocational instruction in the Job Corps (JC) under the assumption that selection into different lengths is based on a rich set of observed covariates and time-invariant factors. We find that the estimated effects on future earnings increase in the length of exposure and that the marginal effects of additional instruction decrease with length of exposure. We also document differences in the estimated effects across demographic groups, which are particularly large between males and females. Finally, our results suggest an important lock-in effect in JC training. No rights reserved. This work was authored as part of the Contributor's official duties as an Employee of the United States Government and is therefore a work of the United States Government. In accordance with 17 U.S.C. 105, no copyright protection is available for such works under U.S. law..

Although economists have long been aware of Jensen's inequality, many econometric applications have neglected an important implication of it: under heteroskedasticity, the parameters of log-linearized models estimated by OLS lead to biased estimates of the true elasticities. We explain why this problem arises and propose an appropriate estimator. Our criticism of conventional practices and the proposed solution extend to a broad range of applications where log-linearized equations are estimated. We develop the argument using one particular illustration, the gravity equation for trade. We find significant differences between estimates obtained with the proposed estimator and those obtained with the traditional method.

Although the double-bound (DB) format for the discrete choice contingent valuation method (CVM) has the benefit of higher efficiency in welfare benefit estimates than the single-bound (SB) discrete choice CVM, it has been subject to criticism due to evidence that some of the responses to the second bid may be inconsistent with the responses to the first bid. As a means to reduce the potential for response bias on the follow-up bid in multiple-bound discrete choice formats such as the DB model while maintaining much of the efficiency gains of the multiple-bound approach, we introduce the one-and-one-half-bound (OOHB) approach and present a real-world application. In a laboratory setting, despite the fact that the OOHB model uses less information than the DB approach, the efficiency gains in moving from SB to OOHB capture a large portion of the gain associated with moving from SB to DB. Utilizing distribution-free seminonparametric estimation techniques on a split-survey data set, our OOHB estimates demonstrated higher consistency with respect to the follow-up data than the DB estimates and were more efficient as well. Hence, OOHB may serve as a viable alternative to the DB format in situations where follow-up response bias may be a concern.

This paper uses three different approaches to investigate whether the declining provision of public capital is a major cause of declining labor productivity. The juxtaposition of approaches removes the variability in estimates due to dissimilar variable definitions and econometric methodologies. Estimates are based on U.S. time-series data and are evaluated by the implied elasticities of substitution, the prediction of labor productivity trends, and the impact of public capital on productivity. As the three approaches yield very different estimates, it will be hard to ever settle the debate about the effect of public capital on private productivity.

Terence C. Mills, Gianluigi Pelloni, Athina Zervoyianni, Unemployment Fluctuations in the United States: Further Tests of the Sectoral-Shifts Hypothesis, The Review of Economics and Statistics, Vol. 77, No. 2 (May, 1995), pp. 294-304