1. Introduction This paper discusses some theoretical implications of recasting parsing and semantic interpretation as a type of inference process which we call incrementalhescription refinement.* It draws upon our recent experience with RUS. a framework for natural language processing developed at BBN and in use in several different natural language systems across the country (for details see ...

In this paper we estimate quadratic food Engel curves with measurement error using the 1994 Ethiopian Urban Household Budget Survey. In sharp contrast to the findings from developed countries, food share is found to increase with expenditure, and only starts to decline after some threshold level of expenditure. We establish the robustness of this inverted-U relationship by estimating a more fle...

Using a sample of male and female workers from the 1992 Employment in Britain survey we estimate a generalised grouped zero-inflated Poisson regression model of employees’ selfreported lateness. Reflecting theoretical predictions from both psychology and economics, lateness is modelled as a function of incentives, the monitoring of and sanctions for lateness within the workplace, job satisfacti...

This paper considers identification and estimation of structural interaction effects in a social interaction model. The model allows unobservables in the group structure, which may be correlated with included regressors. We show that both the endogenous and exogenous interaction effects can be identified if there are sufficient variations in group sizes. We consider the estimation of the model ...

We develop new tests for the coe¢ cient on a time trend in a regression of a variable on a constant and time trend where there is potentially strong serial correlation. This serial correlation can also include a unit root. We obtain tests under two different assumptions on the initial value for the stochastic component of the variable being examined, either this being zero asymptotically (as in...

Frölich (2004) compares the finite sample properties of reweighting and matching estimators of average treatment effects and concludes that reweighting performs far worse than even the simplest matching estimator. We argue that this conclusion is unjustified. Neither approach dominates the other uniformly across data generating processes (DGPs). Expanding on the Frölich’s (2004) analysis, this ...

This paper explores the widely accepted view thatWal-Mart causes significant harm to the traditional, small ‘‘mom and pop’’ business sector of the U.S. economy. We present the first rigorous econometric investigation of this issue by examining the rate of self-employment and the number of small employer establishments using both time series and cross-sectional data. We also examine alternative ...

This article considers quasi-maximum likelihood estimations (QMLE) for two spatial panel data regression models: mixed effects model with spatial errors and transformed mixed effects model (where response and covariates are transformed) with spatial errors. One aim of transformation is to normalize the data, thus the transformed models are more robust with respect to the normality assumption co...

This paper demonstrates a method for estimating treatment effects in spatial tests, utilizing a second control group to measure unexplained spatial phenomena. The technique is implemented on two innovations in Ugandan microfinance, and we measure the ways in which concurrent shocks such as an ebola outbreak and a contentious presidential election altered outcomes differentially across regions. ...

This article provides the first evidence that universities compete directly on price, and that the market for students depends on the proximity of competitors. Exploiting detailed data from private U.S. universities, price competition is tested by introducing geographic proximity into a spatial-autoregressive tuition model. Standard spatial models show that list and net tuition are inversely re...