Estimating learning models from experimental data

We study the statistical properties of three estimation methods for a model of learning that is often fitted to experimental data: quadratic deviation measures without unobserved heterogeneity, and maximum likelihood withand without unobserved heterogeneity. After discussing identification issues, we show that the estimators are consistent and provide their asymptotic distribution. Using Monte Carlo simulations, we show that ignoring unobserved heterogeneity can lead to seriously biased estimations in samples which have the typical length of actual experiments. Better small sample properties areobtained if unobserved heterogeneity is introduced. That is, rather than estimating the parameters for each individual, the individual parameters are considered random variables, and the distribution of those random variables is estimated.

Location models for ceding market share and shrinking services

New location models are presented here for exploring the reduction of facilities in aregion. The first of these models considers firms ceding market share to competitorsunder situations of financial exigency. The goal of this model is to cede the leastmarket share, i.e., retain as much of the customer base as possible while sheddingcostly outlets. The second model considers a firm essentially without competition thatmust shrink it services for economic reasons. This firm is assumed to close outlets sothat the degradation of service is limited. An example is offered within a competitiveenvironment to demonstrate the usefulness of this modeling approach.

Radial measures of public services deficit for regional allocation of public funds

The goal of this paper is to present an optimal resource allocation model for the regional allocation of public service inputs. Theproposed solution leads to maximise the relative public service availability in regions located below the best availability frontier, subject to exogenous budget restrictions and equality ofaccess for equal need criteria (equity-based notion of regional needs). The construction of non-parametric deficit indicators is proposed for public service availability by a novel application of Data Envelopment Analysis (DEA) models, whose results offer advantages for the evaluation and improvement of decentralised public resource allocation systems. The method introduced in this paper has relevance as a resource allocation guide for the majority of services centrally funded by the public sector in a given country, such as health care, basic and higher education, citizen safety, justice, transportation, environmental protection, leisure, culture, housing and city planning, etc.

The impact of firing costs on turnover and unemployment: Evidence from the Colombian labour market reform

Reductions in firing costs are often advocated as a way of increasingthe dynamism of labour markets in both developed and less developed countries. Evidence from Europe and the U.S. on the impact of firing costs has, however, been mixed. Moreover, legislative changes both in Europe and the U.S. have been limited. This paper, instead, examines the impact of the Colombian Labour Market Reform of 1990, which substantially reduced dismissal costs. I estimate the incidence of a reduction in firing costs on worker turnover by exploiting the temporal change in the Colombian labour legislation as well as the variability in coverage between formal and informal sector workers. Using a grouping estimator to control for common aggregate shocks and selection, I find that the exit hazard rates into and out of unemployment increased after the reform by over 1% for formal workers (covered by the legislation) relative to informal workers (uncovered). The increase of the hazards implies a net decrease in unemployment of a third of a percentage point, which accounts for about one quarter of the fall in unemployment during the period of study.

Location models for airline hubs behaving as M/D/c queues

Models are presented for the optimal location of hubs in airline networks, that take into consideration the congestion effects. Hubs, which are the most congested airports, are modeled as M/D/c queuing systems, that is, Poisson arrivals, deterministic service time, and {\em c} servers. A formula is derived for the probability of a number of customers in the system, which is later used to propose a probabilistic constraint. This constraint limits the probability of {\em b} airplanes in queue, to be lesser than a value $\alpha$. Due to the computational complexity of the formulation. The model is solved using a meta-heuristic based on tabu search. Computational experience is presented.

On heuristic and linear models of judgment: Mapping the demand for knowledge

Research on judgment and decision making presents a confusing picture of human abilities. For example, much research has emphasized the dysfunctional aspects of judgmental heuristics, and yet, other findings suggest that these can be highly effective. A further line of research has modeled judgment as resulting from as if linear models. This paper illuminates the distinctions in these approaches by providing a common analytical framework based on the central theoretical premise that understanding human performance requires specifying how characteristics of the decision rules people use interact with the demands of the tasks they face. Our work synthesizes the analytical tools of lens model research with novel methodology developed to specify the effectiveness of heuristics in different environments and allows direct comparisons between the different approaches. We illustrate with both theoretical analyses and simulations. We further link our results to the empirical literature by a meta-analysis of lens model studies and estimate both human andheuristic performance in the same tasks. Our results highlight the trade-off betweenlinear models and heuristics. Whereas the former are cognitively demanding, the latterare simple to use. However, they require knowledge and thus maps of when andwhich heuristic to employ.

The evolution of inequity in the access to health care in Spain: 1987-2001

This paper reports an analysis of the evolution of equity in access to health care in Spain over the period 1987-2001, a time span covering the development of the modern Spanish National Health System. Our measures of access are the probabilities of visiting a doctor, using emergency services and being hospitalised. For these three measures we obtain indices of horizontal inequity from microeconometric models of utilization that exploit the individual information in the Spanish National Health Surveys of 1987 and 2001. We find that by 2001 the system has improved in the sense that differences in income no longer lead to different access given the same level of need. However, the tenure of private health insurance leads to differences in access given the same level of need, and its contribution to inequity has increased over time, both because insurance is more concentrated among the rich and because the elasticity of utilization for the three services has increased too.

The marginal propensity to spend on adult children

We examine how much of an extra dollar of parental lifetime resources willultimately be passed on to adult children in the form of inter vivostransfers and bequests. We infer bequests from the stock of wealth late inlife. We use mortality rates and age specific estimates of the response oftransfers and wealth to permanent income to compute the expected presentdiscounted values of these responses to permanent income. Our estimatesimply parents pass on between 2 and 3 cents out of an extra dollar ofexpected lifetime resources in bequests and about 2 cents in transfers.The estimates increase with parental income and are smaller for nonwhites.They imply that about 15 percent of the effect of parental income onlifetime resources of adult children is through transfers and bequestsand about 85 percent is through the intergenerational correlation inearnings, although these estimates are sensitive to assumptions about theintergenerational earnings correlation, taxes, and the number of children.We compare our estimates to the implications of alternative computablebenchmark models of savings behavior in order to assess the likelyimportance of intended bequests for the wealth/income relationship.

Hierarchical location-allocation models for congested systems

In this paper we address the issue of locating hierarchical facilities in the presence of congestion. Two hierarchical models are presented, where lower level servers attend requests first, and then, some of the served customers are referred to higher level servers. In the first model, the objective is to find the minimum number of servers and theirlocations that will cover a given region with a distance or time standard. The second model is cast as a Maximal Covering Location formulation. A heuristic procedure is then presented together with computational experience. Finally, some extensions of these models that address other types of spatial configurations are offered.

Estimates of patient costs related with population morbidity: Can indirect costs affect the results?

A number of health economics works require patient cost estimates as a basic information input.However the accuracy of cost estimates remains in general unspecified. We propose to investigate howthe allocation of indirect costs or overheads can affect the estimation of patient costs in order to allow forimprovements in the analysis of patient costs estimates. Instead of focusing on the costing method, thispaper proposes to highlight changes in variance explained observed when a methodology is chosen. Wecompare three overhead allocation methods for a specific Spanish population adjusted using the ClinicalRisk Groups (CRG), and we obtain different series of full-cost group estimates. As a result, there aresignificant gains in the proportion of the variance explained, depending upon the methodology used.Furthermore, we find that the global amount of variation explained by risk adjustment models dependsmainly on direct costs and is independent of the level of aggregation used in the classification system.

Employee referrals and the inter-industry wage structure

This paper investigates the role of employee referrals in the labor market.Using an original data set, I find that industries that pay wage premia andhave characteristics associated with high-wage sectors rely mainly on employeereferrals to fill jobs. Moreover, unemployment rates are higher in industries which use employee referrals more extensively. This paper develops an equilibrium matching model which can explain these empirical regularities. Inthis model, the matching process sorts heterogeneous firms and workers into two distinct groups: referrals match "good" jobs to "good" workers, while formalmethods (e.g., newspaper ads and employment agencies) match less-attractive jobs to disadvantaged workers. Thus, well-connected workers who learn quickly aboutjob opportunities use referrals to jump job queues, while those who are less well placed in the labor market search for jobs through formal methods. The split of firms and workers between referrals and formal search is, however, not necessarily efficient. Congestion externalities in referral search imply that unemployment would be closer to the optimal rate if firms and workers 'at themargin' searched formally.

Protective or counter-productive? Labor market institutions and the effect of immigration on EU natives

We estimate the effect of immigrant flows on native employment in WesternEurope, and then ask whether the employment consequences of immigrationvary with institutions that affect labor market flexibility. Reducedflexibility may protect natives from immigrant competition in the nearterm, but our theoretical framework suggests that reduced flexibility islikely to increase the negative impact of immigration on equilibriumemployment. In models without interactions, OLS estimates for a panel ofEuropean countries in the 1980s and 1990s show small, mostly negativeimmigration effects. To reduce bias from the possible endogeneity ofimmigration flows, we use the fact that many immigrants arriving after1991 were refugees from the Balkan wars. An IV strategy based onvariation in the number of immigrants from former Yugoslavia generateslarger though mostly insignificant negative estimates. We then estimatemodels allowing interactions between the employment response toimmigration and institutional characteristics including business entrycosts. These results, limited to the sample of native men, generallysuggest that reduced flexibility increases the negative impact ofimmigration. Many of the estimated interaction terms are significant,and imply a significant negative effect on employment in countrieswith restrictive institutions.

Validation procedures in radiological diagnostic models. Neural network and logistic regression

The objective of this paper is to compare the performance of twopredictive radiological models, logistic regression (LR) and neural network (NN), with five different resampling methods. One hundred and sixty-seven patients with proven calvarial lesions as the only known disease were enrolled. Clinical and CT data were used for LR and NN models. Both models were developed with cross validation, leave-one-out and three different bootstrap algorithms. The final results of each model were compared with error rate and the area under receiver operating characteristic curves (Az). The neural network obtained statistically higher Az than LR with cross validation. The remaining resampling validation methods did not reveal statistically significant differences between LR and NN rules. The neural network classifier performs better than the one based on logistic regression. This advantage is well detected by three-fold cross-validation, but remains unnoticed when leave-one-out or bootstrap algorithms are used.

Job search and asset accumulation under borrowing constraints

In this paper I show how borrowing constraints and job search interact.I fit a dynamic model to data from the National Longitudinal Survey(1979-cohort) and show that borrowing constraints are significant. Agentswith more initial assets and more access to credit attain higher wagesfor several periods after high school graduation. The unemployed maintaintheir consumption by running down their assets, while the employed saveto buffer against future unemployment spells. I also show that, unlikein models with exogenous income streams, unemployment transfers, byallowing agents to attain higher wages do not 'crowd out' but increasesaving.