This paper considers testing problems where several of the standard regularity conditions fail to hold. We consider the case where (i) parameter vectors in the null hypothesis may lie on the boundary of the maintained hypothesis and (ii) there may be a nuisance parameter that appears under the alternative hypothesis, but not under the null. The paper establishes the asymptotic null and local alternative distributions of quasi-likelihood ratio, rescaled quasi-likelihood ratio, Wald, and score tests in this case. The results apply to tests based on a wide variety of extremum estimators and apply to a wide variety of models. Examples treated in the paper are: (i) tests of the null hypothesis of no conditional heteroskedasticity in a GARCH(1, 1) regression model and (ii) tests of the null hypothesis that some random coefficients have variances equal to zero in a random coefficients regression model with (possibly) correlated random coefficients.

This study puts forth stationarity considerations in explaining the observed breakdown between aggregate Discount Window borrowing and the spread between the Federal Funds rate and the discount rate during the post-1987 period. Tests with biweekly data indicate stationarity for adjustment borrowing, but cannot reject the unit root for the spread. The Goodfriend–Dutkowsky dynamic implicit cost formulation can accommodate the contrasting stationarity properties. Structural restrictions are compatible with stationary borrowing and a stationary or near integrated spread. While empirical findings from the static model indicate greater bank reluctance to borrow over time, the dynamic model gives considerably less support.

Properties of instrumental variable estimators are sensitive to the choice of valid instruments, even in large cross-section applications. In this paper we address this problem by deriving simple mean-square error criteria that can be minimized to choose the instrument set. We develop these criteria for two-stage least squares (2SLS), limited information maximum likelihood (LIML), and a bias adjusted version of 2SLS (B2SLS). We give a theoretical derivation of the mean-square error and show optimality. In Monte Carlo experiments we find that the instrument choice generally yields an improvement in performance. Also, in the Angrist and Krueger (1991) returns to education application, when the instrument set is chosen in the way we consider, it turns out that both 2SLS and LIML give similar (large) returns to education.

A half century of empirical work attempting to predict the factor content of trade in goods has failed to bring theory and data into congruence. Our study shows how the Heckscher-Ohlin-Vanek theory, when modified to permit technical differences, a breakdown in factor price equalization, the existence of nontraded goods, and costs of trade, is consistent with data from ten OECD countries and a rest-of-world aggregate. (JEL F1, F11, D5)

Building on the Dechow et al. (1998) model of the accrual process, this study investigates the role of accruals in predicting future cash flows. The model shows that each accrual component reflects different information relating to future cash flows; aggregate earnings masks this information. As predicted, disaggregating accruals into major components—change in accounts receivable, change in accounts payable, change in inventory, depreciation, amortization, and other accruals—significantly enhances predictive ability. Each accrual component, including depreciation and amortization, is significant with the predicted sign in predicting future cash flows, incremental to current cash flow. The cash flow and accrual components of current earnings have substantially more predictive ability for future cash flows than several lags of aggregate earnings. The inferences are robust to alternative specifications, including controlling for operating cash cycle and industry membership.

This paper provides empirical evidence about the effect of unearned income on earnings, consumption, and savings. Using an original survey of people playing the lottery in Massachusetts in the mid-1980's, we analyze the effects of the magnitude of lottery prizes on economic behavior. The critical assumption is that among lottery winners the magnitude of the prize is randomly assigned. We find that unearned income reduces labor earnings, with a marginal propensity to consume leisure of approximately 11 percent, with larger effects for individuals between 55 and 65 years old. After receiving about half their prize, individuals saved about 16 percent. (JEL C81, D12, E21, J22, J26)

Longevity Expectations and Death: Can People Predict Their Own Demise? by V. Kerry Smith, Donald H. Taylor and Frank A. Sloan. Published in volume 91, issue 4, pages 1126-1134 of American Economic Review, September 2001

This paper reports the first effort to use data to evaluate how new information, acquired through exogenous health shocks, affects people's longevity expectations. We find that smokers react differently to health shocks than do those who quit smoking or never smoked. These differences, together with insights from qualitative research conducted along with the statistical analysis, suggest specific changes in the health warnings used to reduce smoking. Our specific focus is on how current smokers responded to health information in comparison to former smokers and nonsmokers. The three groups use significantly different updating rules to revise their assessments about longevity. The most significant finding of our study documents that smokers differ from persons who do not smoke in how information influences their personal longevity expectations. When smokers experience smoking-related health shocks, they interpret this information as reducing their chances of living to age 75 or more. Our estimated models imply smokers update their longevity expectations more dramatically than either former smokers or those who never smoked. Smokers are thus assigning a larger risk equivalent to these shocks. They do not react comparably to general health shocks, implying that specific information about smoking-related health events is most likely to cause them to update beliefs. It remains to be evaluated whether messages can be designed that focus on the link between smoking and health outcomes in ways that will have comparable effects on smokers' risk perceptions.

This study reports the results of three experiments that examine how preferences for wealth and honesty affect managerial reporting. We find that subjects often sacrifice wealth to make honest or partially honest reports, and they generally do not lie more as the payoff to lying increases. We also find less honesty under a contract that provides a smaller share of the total surplus to the manager than under one that provides a larger share, suggesting that the extent of honesty may depend on how the surplus is divided between the manager and the firm. The optimal agency contract yields more firm profit than a contract that relies exclusively on honest reporting. However, a modified version of the optimal agency contract, which makes use of subjects' preferences for honest reporting, yields the highest firm profit. These results suggest that firms may be able to design more profitable employment contracts than those identified by conventional economic analysis.