SOME PRACTICAL GUIDANCE FOR THE IMPLEMENTATION OF PROPENSITY SCORE MATCHING

Abstract Propensity score matching (PSM) has become a popular approach to estimate causal treatment effects. It is widely applied when evaluating labour market policies, but empirical examples can be found in very diverse fields of study. Once the researcher has decided to use PSM, he is confronted with a lot of questions regarding its implementation. To begin with, a first decision has to be made concerning the estimation of the propensity score. Following that one has to decide which matching algorithm to choose and determine the region of common support. Subsequently, the matching quality has to be assessed and treatment effects and their standard errors have to be estimated. Furthermore, questions like ‘what to do if there is choice‐based sampling?’ or ‘when to measure effects?’ can be important in empirical studies. Finally, one might also want to test the sensitivity of estimated treatment effects with respect to unobserved heterogeneity or failure of the common support condition. Each implementation step involves a lot of decisions and different approaches can be thought of. The aim of this paper is to discuss these implementation issues and give some guidance to researchers who want to use PSM for evaluation purposes.     

A comparison of treatment effects estimators using a structural model of AMI treatment choices and severity of illness information from hospital charts

We compare the performance of various matching estimators using a novel approach that is feasible in the absence of experimental data. We estimate a structural model of hospital choices and catheterization for Medicare heart attack victims using hospital chart data on patient heterogeneity. With the estimated structural parameters, we simulate data for which the treatment effect is known. We find that as measures of individual heterogeneity are added to the controls, matching estimators perform well. However, the estimators do a poor job recovering the true treatment effect when measures of individual heterogeneity are unavailable. Copyright © 2010 John Wiley & Sons, Ltd.     

Time and causality: A Monte Carlo assessment of the timing-of-events approach

We present new Monte Carlo evidence regarding the feasibility of separating causality from selection within non-experimental duration data, by means of the non-parametric maximum likelihood estimator (NPMLE). Key findings are: (i) the NPMLE is extremely reliable, and it accurately separates the causal effects of treatment and duration dependence from sorting effects, almost regardless of the true unobserved heterogeneity distribution; (ii) the NPMLE is normally distributed, and standard errors can be computed directly from the optimally selected model; and (iii) unjustified restrictions on the heterogeneity distribution, e.g., in terms of a pre-specified number of support points, may cause substantial bias.     

Implementation of voting operators

We show that a theory of implementation can be developed in the Aizerman–Aleskerov framework, capturing the main ideas regarding Nash implementation in the Arrovian case. In fact we obtain the counterparts of the results of [Maskin, E., 1977. Nash Equilibrium and Welfare Optimality. MIT, Mimeo] and [Moore, J., Repullo, R., 1990. Nash implementation. A full characterization. Econometrica 58, 1038–1100] in the new framework.     

How to Regulate Heterogeneous Hospitals?

In many areas of health care financing, there is controversy over the sources of cost variability and about the respective roles of inefficiency versus legitimate heterogeneity. This paper proposes a payment system that creates incentives to increase hospital efficiency when hospitals are heterogeneous, without reducing the quality of care. We consider an extension of Shleifer's yardstick competition model and apply an econometric approach to identify and evaluate observable and unobservable sources of cost heterogeneity. Moral hazard can be seen as the result of two components: long‐term moral hazard (hospital management can be permanently inefficient) and transitory moral hazard. The latter is linked to the manager's transitory cost‐reducing effort. For instance, he or she can be more or less rigorous each year when bargaining prices for supplies delivered to the hospital by outside firms. The use of a three‐dimensional nested database makes it possible to identify transitory moral hazard and to estimate its effect on hospital cost variability. Econometric estimates are performed on a sample of 7,314 stays for acute myocardial infarction observed in 36 French public hospitals over the period 1994–1997. We obtain two alternative payment systems. The first takes all unobservable hospital heterogeneity into account, provided that it is time invariant, whereas the second ignores unobservable heterogeneity. Simulations show that substantial budget savings—at least 20%—can be expected from the implementation of such payment rules. The first method of payment has the great advantage of reimbursing high‐quality care. It leads to substantial potential savings because it provides incentives to reduce costs linked to transitory moral hazard, whose influence on cost variability is far from negligible. This payment rule could be extended to other areas of health care financing, such as Adjusted Average Per Capita Cost to calculate Medicare Managed Care reimbursements in the United States.     

An empirical assessment of assortative matching in the labor market

In labor markets with worker and firm heterogeneity, the matching between firms and workers may be assortative, meaning that the most productive workers and firms team up. We investigate this with longitudinal population-wide matched employer–employee data from Portugal. Using panel data methods, we quantify a firm-specific productivity term for each firm, and we relate this to the skill distribution of workers in the firm. We find that there is positive assortative matching, in particular among long-lived firms. Using skill-specific estimates of an index of search frictions, we find that the results can only to a small extent be explained by heterogeneity of search frictions across worker skill groups.     

A latent class poisson regression model for heterogeneous count data

In this paper an approach is developed that accommodates heterogeneity in Poisson regression models for count data. The model developed assumes that heterogeneity arises from a distribution of both the intercept and the coefficients of the explanatory variables. We assume that the mixing distribution is discrete, resulting in a finite mixture model formulation. An EM algorithm for estimation is described, and the algorithm is applied to data on customer purchases of books offered through direct mail. Our model is compared empirically to a number of other approaches that deal with heterogeneity in Poisson regression models.     

Endogenous job destructions and the distribution of wages

This paper considers a matching model with both idiosyncratic productivity shocks that hit jobs at random and heterogeneity of workers according to ex ante unobservable abilities. We argue that firms' decisions about reservation productivity can help explain the shape of wage distributions. This is shown from numerical experiments, calibrated to French data, by considering alternative ranges of productivity shocks.     

Career Tournaments

This paper studies a model of career concerns where (besides agents having different ability levels) principals are heterogeneous. Two types of heterogeneity are considered: principals can differ in their productivities, and in the visibility of the agents working for them. Productivity differences result in convex wage profiles and generate assortative matching between principals and agents. Visibility differences, on the other hand, affect the mobility and career prospects of agents, and can generate nonassortative matching. We characterize the equilibrium in this market, derive implications for the behavior of agents, and examine the incentives to invest in increasing productivity and visibility.     

A Test of the Conditional Independence Assumption in Sample Selection Models

Identification in most sample selection models depends on the independence of the regressors and the error terms conditional on the selection probability. All quantile and mean functions are parallel in these models; this implies that quantile estimators cannot reveal any—per assumption non‐existing—heterogeneity. Quantile estimators are nevertheless useful for testing the conditional independence assumption because they are consistent under the null hypothesis. We propose tests of the Kolmogorov–Smirnov type based on the conditional quantile regression process. Monte Carlo simulations show that their size is satisfactory and their power sufficient to detect deviations under plausible data‐generating processes. We apply our procedures to female wage data from the 2011 Current Population Survey and show that homogeneity is clearly rejected. Copyright © 2015 John Wiley & Sons, Ltd.     

The Hiring Function Reconsidered: on Closing the Circle

This paper investigates the matching of job searchers with vacant jobs: a key component of the dynamics of worker reallocation in the labour market. The job searchers may be unemployed, employed or not in the labour force and we estimate matching or hiring functions including all three groups. We show that previous studies, which ignore both employed job seekers and unemployed job seekers who are considered to be out of the labour force, produce biased estimates of the coefficients of interest. By considering only unemployment outflows into jobs and ignoring interdependencies with other flows, these studies overlook an important aspect of job matching. Our estimates on Australian data support a more general approach and produce models that dominate those proposed previously. We find that concentrating on the aggregate matching function alone does not reveal the full extent of the interaction across job searchers. Indeed, we find that job searchers from the three groups do not receive a fair share of hires: there appears to be segmentation of hiring opportunities which may be explained by a form of ranking of applicants. Together these results demonstrate that the disaggregate worker flows and their interdependence are key features on the labour market and should be included in studies of the hiring process.     

Matching and semi-parametric IV estimation, a distance-based measure of migration, and the wages of young men

Our paper estimates the effect of US internal migration on wage growth for young men between their first and second job. Our analysis of migration extends previous research by: (i) exploiting the distance-based measures of migration in the National Longitudinal Surveys of Youth 1979 (NLSY79); (ii) allowing the effect of migration to differ by schooling level and (iii) using propensity score matching to estimate the average treatment effect on the treated (ATET) for movers and (iv) using local average treatment effect (LATE) estimators with covariates to estimate the average treatment effect (ATE) and ATET for compliers.We believe the Conditional Independence Assumption (CIA) is reasonable for our matching estimators since the NLSY79 provides a relatively rich array of variables on which to match. Our matching methods are based on local linear, local cubic, and local linear ridge regressions. Local linear and local ridge regression matching produce relatively similar point estimates and standard errors, while local cubic regression matching badly over-fits the data and provides very noisy estimates.We use the bootstrap to calculate standard errors. Since the validity of the bootstrap has not been investigated for the matching estimators we use, and has been shown to be invalid for nearest neighbor matching estimators, we conduct a Monte Carlo study on the appropriateness of using the bootstrap to calculate standard errors for local linear regression matching. The data generating processes in our Monte Carlo study are relatively rich and calibrated to match our empirical models or to test the sensitivity of our results to the choice of parameter values. The estimated standard errors from the bootstrap are very close to those from the Monte Carlo experiments, which lends support to our using the bootstrap to calculate standard errors in our setting.From the matching estimators we find a significant positive effect of migration on the wage growth of college graduates, and a marginally significant negative effect for high school dropouts. We do not find any significant effects for other educational groups or for the overall sample. Our results are generally robust to changes in the model specification and changes in our distance-based measure of migration. We find that better data matters; if we use a measure of migration based on moving across county lines, we overstate the number of moves, while if we use a measure based on moving across state lines, we understate the number of moves. Further, using either the county or state measures leads to much less precise estimates.We also consider semi-parametric LATE estimators with covariates (Frölich 2007), using two sets of instrumental variables. We precisely estimate the proportion of compliers in our data, but because we have a small number of compliers, we cannot obtain precise LATE estimates.     

具有自动撮合与视频通信的高校就业网站设计

文章针对传统就业网站功能单一、服务起点低、没有充分发挥网络交互通信作用的弊端,提出了一个具有撮合算法和视频通信功能的新型招应聘网络平台的设计,文章介绍了系统的结构、基于模糊向量的撮合算法以及基于对等通信模型的视频功能实现,对同类研究和设计有一定参考作用。     

Zero-intelligence agents looking for a job

We study a simple agent-based model of a decentralized matching market game in which agents (workers or job seekers) make proposals to other agents (firms) in order to be matched to a position within the firm. The aggregate result of agents interactions can be summarised in the form of a Beveridge curve, which determines the relationship between unmatched agents, unemployed job seekers and vacancies in firms. We open the black box of matching technology, by modelling how agents behave (make proposals) according to their information perception. We observe more efficient results—in the form of a downward shift of the Beverage curve in the case of simple zero-intelligent agents. Our comparative statics indicate that market conditions, such as the heterogeneity of agents’ preferences, will also shift the Beveridge curve downwards. Moreover, market thickness affects movement along the Beverage curve. Movement right-down along the curve if there is an increasing number of agents compared to positions within firms. Furthermore, we show that frictions in re-matching, such as commitment to a match, could be another factor shifting the Beveridge curve toward the origin.     

STRATEGIES OF EXPANSION AND ORGANIZATIONAL DEVELOPMENTS IN EUROPEAN AND AMERICAN FIRMS 1

Summary The pace of change of the last ten years has challenged the traditional management processes to the point that the key organizational question is no longer to fit new strategies into available organizational patterns but instead to fit the organizational structure to the kind of problems to be faced. However, the intricacy of strategic and structural difficulties is often such that the appraisal of organizational inadequacies may prove difficult: symptoms may be wrongly considered as problems which may in turn be over-or understated so that corrective measures may well be inappropriate. This is why the saying‘When in trouble, reoganize!’is of little practicability. The lack of organizational planning in framing responses to structural problems will more and more make adjustments necessary. In our attempts to define the nature of organizational changes which are likely to occur within business firms in the future, we shall isolate the stragetic decisions related to corporate expansion on a transnational, trans-continental and eventually on an intercontinental scale. We shall start by analysing present organizational patterns — as they reflect decisions made in the past regarding current strategies — and we shall try to outline some possible structural developments as they may be required by the implementation of future strategies.     

Matching in local labor markets: a stochastic frontier approach

This paper investigates the technical efficiency of labor market matching from a stochastic frontier approach. The true fixed-effects model (Greene J Prod Anal 23:7–32, 2005a; J Econom 126:269–303, 2005b) is utilised in order to separate cross-sectional heterogeneity from inefficiency, and inefficiency terms are modelled following Battese and Coelli (Empir Econ 20:325–332, 1995). The data set consists of almost 17,000 observations from Local Labor Offices (LLOs) in Finland. According to the results, there are notable differences in matching efficiency between regions, and these differences contribute significantly to the number of filled vacancies. If all regions were as efficient as the most efficient one, the number of total matches per month would increase by over 23%. The heterogeneity of the job-seeker stock is an important determinant of matching efficiency: the weight of the composition of the job-seeker stock in the inefficiency terms is on average 85%.

A modified deferred acceptance algorithm for many-to-one matching markets with externalities among firms

We consider a many-to-one matching market with externalities among firms where each firm’s preferences satisfy substitutability, increasing choice and no external effect by unchosen workers, which are defined by  Bando (2012). We first illustrate that a sequential version of the deferred acceptance (DA) algorithm with worker-proposing may not find a worker-optimal quasi stable matching. Then, we provide a modified DA algorithm in which (i) each worker simultaneously proposes to his most preferred firm that has not rejected him and (ii) each firm chooses its acceptable workers from the cumulative set of workers who have ever proposed to it, assuming that the other workers proposing to its rival firms are hired. We show that this algorithm finds a worker-optimal quasi stable matching. We also show that this algorithm can be generalized into a fixed point algorithm.     

Are non‐binding contracts really not worth the paper?

We experimentally investigate behavior in sequential one‐shot transactions which are governed by non‐binding contracts. In a second, incomplete information treatment, contracts are binding for some players. While according to traditional game‐theoretical analysis no trade is expected in the first treatment, full trade should result in the latter. However, we find that trade is even higher in the non‐binding contract treatment. On the one hand, non‐binding contracts—although they are cheap talk—do guide behavior, especially at the beginning of a business relationship, while reciprocal reactions prevail later on. On the other hand, in the treatment with binding contracts cooperative behavior appears to be crowded out. Copyright © 2005 John Wiley & Sons, Ltd.     

Heterogeneous impacts in PROGRESA

We investigate impact heterogeneity using data from the experimental evaluation of the Mexican conditional cash transfer program PROGRESA. We build upon, and extend Heckman, Smith and Clements [Heckman, J., Smith, J., Clements, N., 1997. Making the most out of programme evaluations and social experiments: Accounting for heterogeneity in programme impacts. Review of Economic Studies 64, 487–535] and recent studies of quantile treatment effects and random coefficient models. We find strong evidence of systematic (i.e. subgroup) variation in impacts in PROGRESA and modest evidence of heterogeneous impacts conditional on the systematic impacts. Our paper concludes with a discussion of the policy relevance of our findings and of heterogeneous impacts more generally.