基于行政大数据的失业率估计:以某四百万人口城市为例

我国城镇登记失业率指标稳定在4%左右,难以较为准确反映就业动态;而劳动力调查样本量有限,城镇调查失业率对省以下各级行政区域代表性不足。本文将针对大数据的机器学习算法与针对传统统计数据的核算思想结合起来,基于某四百万人口城市2016—2018年的全样本行政大数据,利用机器学习算法,对每个城镇居民每个月的就业状态进行预测,再利用统计核算方法,估计出该城市的失业率。在个人层面,本文的模型在样本外测试集上的准确率达到96.7%。经过统计核算加总,本文估计的当地失业率在合理区间范围内,并表现出明显的周期性特征,对就业形势动态变化的刻画明显优于当地一年发布一次的登记失业率数据。本文基于个人层面的预测结果,进一步探讨了当地失业人口 的性别与文化程度特征,以及再就业的时间规律。本文针对如何使用行政大数据辅助经济决策提出了新的范式,对大数据时代如何理解经济与制定政策具有参考意义。     

Time Series Evidence of Unemployment Flows: The Sample Period Matters

Worker flows to and from unemployment occur at all stages of the U.S. business cycle. The flows' dynamics affect the dynamics of the unemployment rate. This article studies the dynamics of unemployment flows and the unemployment rate and explores and documents changes over time. Rigorous diagnostic tests are used to investigate the structural stability of projectional relationships involving flows and the unemployment rate.     

Hierarchical Bayes estimation of spatial statistics for rates

The U.S. Bureau of Labour Statistics publishes monthly unemployment rate estimates for its 50 states, the District of Columbia, and all counties, under Current Population Survey. However, the unemployment rate estimates for some states are unreliable due to low sample sizes in these states. Datta et al. (1999) proposed a hierarchical Bayes (HB) method using a time series generalization of a widely used cross-sectional model in small-area estimation. However, the geographical variation is also likely to be important. To have an efficient model, a comprehensive mixed normal model that accounts for the spatial and temporal effects is considered. A HB approach using Markov chain Monte Carlo is used for the analysis of the U.S. state-level unemployment rate estimates for January 2004-December 2007. The sensitivity of such type of analysis to prior assumptions in the Gaussian context is also studied.     

Producing Monthly Estimates of Unemployment and Employment According to the International Labour Office Definition

Monthly unemployment statistics are available in Britain from a monthly count of the number of people claiming unemployment-related benefits. There has been considerable debate on the appropriateness of this measure. Unemployment and employment statistics are available quarterly from the Labour Force Survey (LFS), using International Labour Office (ILO) definitions. In this paper various options for producing monthly unemployment estimates according to the ILO definition are examined. Methods considered are a monthly LFS, calculating rolling averages from the quarterly LFS, and methods which combine LFS and claimant count data. It is proposed that a monthly LFS of 60 000 households be introduced which can produce monthly estimates of total unemployment and more detailed estimates quarterly. Such a survey would also fill an important gap by providing monthly employment statistics which are needed to provide a complete picture of the labour market.     

Bias in Multinomial Models

This paper investigates bias in parameter estimates and residual diagnostics for parametric multinomial models by considering the effect of deleting a cell. In particular, it describes the average changes in the standardized residuals and maximum likelihood estimates resulting from conditioning on the given cells. These changes suggest how individual cell observations affect biases. Emphasis is placed on the role of individual cell observations in determining bias and on how bias affects standard diagnostic methods. Examples from genetics and log–linear models are considered. Numerical results show that conditioning on an influential cell results in substantial changes in biases.     

Measuring the Experience-Productivity Relationship: The Case of Major League Baseball

This article develops and estimates a model of the typical major league baseball player's lifetime batting average profile. The batting average profile can be thought of as an experience-productivity relationship. The estimation techniques are those suggested by MaCurdy (1981 a, b) for estimating model parameters when using panel data, including the parameters of the error structure. The method could be used to estimate the relationship between experience and productivity in any situation where a measure of individual output is available. The results include estimates of the peak of a baseball player's career and the rate at which batting averages change with age.     

Survey non-response and the duration of unemployment

Summary.  Social surveys are often used to estimate unemployment duration distributions. Survey non-response may then cause a bias. We study this by using a data set that combines survey information of individual workers with administrative records of the same workers. The latter provide information on durations of unemployment and personal characteristics of all survey respondents and non-respondents. We develop a method to distinguish empirically between two explanations for a bias in results based on only survey data: selectivity due to related unobserved determinants of durations of unemployment and non-response and a causal effect of a job exit on non-response. The latter may occur even in fully homogeneous populations. The methodology exploits variation in the timing of the duration outcome relative to the survey moment. The results show evidence for both explanations. We discuss implications for standard methods to deal with non-response bias.     

Unemployment duration and the length of entitlement periods for unemployment benefits: do the IAB employment subsample and the German Socio-Economic Panel yield the same results?*

Summary: We compare information on the length of unemployment spells contained in the IAB employment subsample (IABS) and in the German Socio-Economic Panel (GSOEP). Due to the lack of information on registered unemployment in the IABS, we use two proxies of unemployment in the IABS as introduced by Fitzenberger/Wilke (2004). The first proxy comprises all periods of nonemployment after an employment spell which contain at least one period with unemployment compensation transfers. The second proxy includes all episodes between two employment spells during which an individual continuously received unemployment benefits. Estimation of standard duration models indicates that conclusions drawn from the IABS and the GSOEP differ in many cases. While the GSOEP suggests that the hazard rate has a maximum at about 12 months of unemployment, the IABS results suggest that this maximum is at about 20 months. Contrary to our GSOEP results and contrary to many results based on the GSOEP found in the literature, we find a statistically significant association between longer maximum entitlement periods of unemployment benefits (‘Arbeitslosengeld’) and longer unemployment durations for men in the IABS. The results for women do not show such clear patterns. The large sample size of the IABS also allows to trace out statistically significant effects of characteristics such as regional and industry indicators, which is generally not possible in the relatively small GSOEP. * Acknowledgements: We would like to thank the editors of this special issue, Joachim M?ller and Bernd Fitzenberger, two anonymous referees, the participants of the ‘Statistische Woche 2004’ in Frankfurt (in particular Reinhard Hujer, Olaf Hübler and Gerd Ronning), seminar participants at the ZEW Mannheim (especially Fran?ois Laisney and Alexander Spermann) and Jennifer Hunt for their many helpful comments and suggestions. All remaining errors are our own. Financial support of the Deutsche Forschungsgemeinschaft (DFG) through the research project ‘Microeconometric modelling of unemployment durations under consideration of the macroeconomic situation’ is gratefully acknowledged. The data used in this paper were made available by the Institute for Employment Research (IAB) at the Federal Labour Office of Germany, Nürnberg, and the German Socio Economic Panel Study (GSOEP) at the German Institute for Economic Research (DIW), Berlin.     

Modelling excess mortality of the unemployed: choice of scale and extra-Poisson variability

"Occupational mortality and morbidity is usually studied via standardized mortality (or morbidity) ratios, with little attention to the basic fit of the implicit underlying proportional hazards model. This paper presents a case study on unemployment and mortality, based on the complete Danish male population aged 20-64 years at the 1970 census. The effect of unemployment on the age-specific mortality rate is intermediate between additive and multiplicative and was fitted well by an additive effect on the square root of the mortality. The paper discusses and illustrates whether finer stratification or random residual variation ('frailty') is to be preferred for obtaining a statistically satisfactory fit."     

Measuring the Influence of Unemployment Insurance on Unemployment Experiences

This article investigates the comprehensive effects of unemployment insurance (UI) policies on the amount of time and unemployment that individuals report between jobs. The econometric model jointly determines the effects of UI on the lengths of nonemployment spells, the classification of these spells as unemployment, and the likelihood of collecting program benefits. The model carefully attempts to isolate variation in UI benefits attributable to differences in generosity across programs to avoid biases in estimating policy effects induced by other contaminating sources of benefit variation. Using data on men from the National Longitudinal Survey of Youth, the empirical results find (a) UI recipients typically experience longer spells between jobs, at least up to the exhaustion of UI benefits, and report substantially larger fractions of these spells as unemployment; (b) weekly benefit amounts exert no significant influence on the likelihood of UI recipiency, on the length of spells between jobs, or on the fraction of these spells classified as unemployment; and (c) increases in weeks of UI eligibility raise the likelihood of UI collection and lengthen the number of weeks of unemployment between jobs by inducing long spells to become longer and not by altering short-duration behavior.     

Monte Carlo integration: An application to labour market states and transitions

This paper is concerned with the application of simulation estimation methods to micro-econometric labour market models. Based on a multi-period probit model for direct job changes and unemployment, estimators for the likelihood of individual employment histories are obtained by Monte Carlo integration and employed in a standard ML-procedure. The results for West German panel data suggest that dynamic effects are largely prevalent on labour markets and that in particular, past unemployment has drastic negative effects on future employment chances. Further, there are no indications that foreigners have a different labour market performance, nor that they are crowding natives out into unemployment.     

The effect of the abolition of capital punishment upon homicides in Great Britain:an application of intervention analysis

In this paper we examine the effect of the abolition of capital punishment on the homicide rate in Great Britain. Previous studies have used econometric techniques to determine whether the death penalty exerts a significant deterrent effect. However, econometric studies have been severely criticized for failing to provide robust results. The method of investigation used in this paper is that of intervention analysis. Some questions are raised about the proper way to measure the intervention effect in integrated models subject to trend changes. However, for both England and Wales, and for Scotland, a statistically significant intervention effect is consistently detected, although the extent of the effect is difficult to determine precisely. We also examine the effect of several ‘social stress indicators' (e.g. suicide and unemployment rates and violent crime) upon the homicide rate.     

Dynamic Asymmetries in U.S. Unemployment

We examine dynamic asymmetries in U.S. unemployment using nonlinear time series models and Bayesian methods. We find strong statistical evidence in favor of a two-regime threshold auto-regressive model. Empirical results indicate that, once we take into account both parameter and model uncertainty, there are economically interesting asymmetries in the unemployment rate. One finding of particular interest is that shocks that lower the unemployment rate tend to have a smaller effect than shocks that raise the unemployment rate. This finding is consistent with unemployment rises being sudden and falls gradual.     

A spatial analysis on Italian unemployment differences

Using spatial econometric models, this paper focuses attention on the spatial structure of provincial unemployment disparities of Italian provinces for the year 2003. On the basis of findings from the economic literature and of the available socio-economic data, various model specifications including supply- and demand-side variables are tested. Further we use ESDA analysis as equivalent to integration analysis on time series; therefore it is applied on each variable, dependent and independent, involved in the statistical model. The suggestions of ESDA lead us to the most adequate statistical model, which estimates indicate that there is a significant degree of neighbouring effect (i.e. positive spatial correlation) among labour markets at the provincial level in Italy; this effect is present notwithstanding we controlled for local characteristics. The unemployment shows a polarized spatial pattern that is strongly connected to labour demand and to a much lesser extent to the share of young population and economic structural composition.     

Estimation of a linear model under microaggregation by individual ranking

Summary Microaggregation by individual ranking is one of themost commonly applied disclosure control techniques for continuous microdata. The paper studies the effect of microaggregation by individual ranking on the least squares estimation of a multiple linear regression model. It is shown that the traditional least squares estimates are asymptotically unbiased. Moreover, the least squares estimates asymptotically have the same variances as the least squares estimates based on the original (non-aggregated) data. Thus, asymptotically, microaggregation by individual ranking does not result in a loss of efficiency in the least squares estimation of a multiple linear regression model. I thank Hans Schneeweiss for very helpful discussions and comments. Financial support from the Deutsche Forschungsgemeinschaft (German Science Foundation) is gratefully acknowledged.     

Sentencing convicted felons in the United States: a Bayesian analysis using multilevel covariates

Imprisonment levels vary widely across the United States, with some state imprisonment rates six times higher than others. Imposition of prison sentences also varies between counties within states, with previous research suggesting that covariates such as crime rate, unemployment level, racial composition, political conservatism, geographic region, and sentencing policies account for some of this variation. Other studies, using court data on individual felons, demonstrate how type of offense, demographics, criminal history, and case characteristics affect sentence severity. This article considers the effects of both county-level and individual-level covariates on whether a convicted felon receives a prison sentence rather than a jail or non-custodial sentence. We analyze felony court case processing data from May 1998 for 39 of the nation's most populous urban counties using a Bayesian hierarchical logistic regression model. By adopting a Bayesian approach, we are able to overcome a number of challenges. The model allows individual-level effects to vary by county, but relates these effects across counties using county-level covariates. We account for missing data using imputation via additional Gibbs sampling steps when estimating the model. Finally, we use posterior samples to construct novel predictor effect plots to aid communication of results to criminal justice policy-makers.     

Keeping Time: An Analysis of Errors in the Measurement of Unemployment Duration

This study uses longitudinal microdata from the Current Population Survey to examine the consistency with which unemployment duration is reported. An estimate of measurement error is proposed and analyzed. A significant negative relationship is found between the initial month's unemployment duration and the reported change between surveys. On average, persons who begin spells of unemployment in the second month report durations greater than the actual time elapsed. Overall, the average duration of unemployment in this sample may be overstated by at least two weeks. A brief discussion of some implications for empirical labor market research concludes the article.     

Influence of light level on the incidence of road casualties and the predicted effect of changing 'summertime'

Previous studies of the apparent influence of daylight level and hour changes on the incidence of road casualties are reviewed and refined, by analysis of official databases for Great Britain (1969–1973 and 1985–1994) and the USA (1991–1995). New statistical methods, based on precisely computed altitudes of the sun for each accident location, are used to model casualty frequencies aggregated by week and hour of day, and locally evaluated associations between individual casualty incidence and solar altitude. Estimates of the altitude factor are interpreted causally to give counterfactual estimates of the effect of different clock time schedules on countrywide casualty numbers.     

对改进中国失业统计方法的探讨

失业问题是人类进入工业化以来一直困扰世界各国的经济和社会问题 ,也是与债务、通货膨胀并列的世界三大难题之一。中国在步入市场经济的过程中 ,同样不可避免地遇到了失业问题的困扰。目前在中国失业现象已普遍存在 ,并且成为影响社会稳定与经济发展的一个非常不利的因素 ,因而引起了社会各界的高度重视。但长期以来 ,中国一直否认社会主义社会有失业问题存在 ,导致国家失业统计纪录的真空。直到 1 993年有关统计部门才将原来所谓的“待业人员”改称为“失业人员” ,这虽然是一个历史性的进步 ,但在失业统计的理论方法以及具体操作环节上却…     

Household Saving in the United States: Measurement and Behavior

Official measures of U.S. personal saving incorporate conceptual errors relating to the measurement of both income and consumption. This article identifies the errors, computes an adjusted personal saving rate to correct the errors, and estimates equations explaining personal saving. The estimates support the adjustments, and the adjusted series has more pronounced movements than the official series, the movements being largely related to changes in income and wealth.