Population forecast accuracy: does the choice of summary measure of error matter?

Population projections are judged primarily by their accuracy. The most commonly used measure for the precision component of accuracy is the mean absolute percent error (MAPE). Recently, the MAPE has been criticized for overstating forecast error and other error measures have been proposed. This study compares the MAPE with two alternative measures of forecast error, the Median APE and an M-estimator. In addition, the paper also investigates forecast bias. The analysis extends previous studies of forecast error by examining a wide range of trend extrapolation techniques using a dataset that spans a century for a large sample of counties in the US. The main objective is to determine whether the choice of summary measure of error makes a difference from a practitioner’s standpoint. The paper finds that the MAPE indeed produces error values that exceed the robust measures. However, except for situations where extreme outliers rendered the MAPE meaningless, and which are rare in real world applications, there was not a single instance where using an alternative summary measure of error would have led to a fundamentally different evaluation of the projections. Moreover, where differences existed, it was not always clear that the values and patterns provided by the robust measures were necessarily more correct than those obtained with the MAPE. While research into refinements and alternatives to the MAPE and mean algebraic percent error are worthwhile, consideration of additional evaluation procedures that go beyond a single criterion might provide more benefits to producers and users of population forecasts.

Identification of Marginal Effects in Nonseparable Models Without Monotonicity

Nonseparable models do not impose any type of additivity between the unobserved part and the observable regressors, and are therefore ideal for many economic applications. To identify these models using the entire joint distribution of the data as summarized in regression quantiles, monotonicity in unobservables has frequently been assumed. This paper establishes that in the absence of monotonicity, the quantiles identify local average structural derivatives of nonseparable models.     

Die Professionalisierung der Professionalisierer?

There is a general agreement that central roles in organizations like manager and consultants cannot be professionalized. However, the development of coaching and supervision seem to show a slightly different picture. Because of their focus on persons the pressure to professionalize seems to be comparable to service providers in law, medicine, religion and education. This article reconstructs the professionalization pressure on coaching and supervision and analyzes their professionalization process.     

Events per variable for risk differences and relative risks using pseudo-observations

A method based on pseudo-observations has been proposed for direct regression modeling of functionals of interest with right-censored data, including the survival function, the restricted mean and the cumulative incidence function in competing risks. The models, once the pseudo-observations have been computed, can be fitted using standard generalized estimating equation software. Regression models can however yield problematic results if the number of covariates is large in relation to the number of events observed. Guidelines of events per variable are often used in practice. These rules of thumb for the number of events per variable have primarily been established based on simulation studies for the logistic regression model and Cox regression model. In this paper we conduct a simulation study to examine the small sample behavior of the pseudo-observation method to estimate risk differences and relative risks for right-censored data. We investigate how coverage probabilities and relative bias of the pseudo-observation estimator interact with sample size, number of variables and average number of events per variable.     

Common consequence effects in pricing and choice

This paper presents an experimental study of common consequence effects in binary choice, willingness-to-pay (WTP) elicitation, and willingness-to-accept (WTA) elicitation. We find strong evidence in favor of the fanning out hypothesis (Machina, Econometrica 50:277–323, 1982) for both WTP and WTA. In contrast, the choice data do not show a clear pattern of violations in the absence of certainty effects. Our results underline the relevance of differences between pricing and choice tasks, and their implications for models of decision making under risk.     

You can go your own way! The long‐term effectiveness of a self‐employment programme for welfare recipients in Germany

In this study, we examined the long‐term effects of supporting welfare recipients by means of a subsidy to start their own business. We used administrative data and propensity score matching to estimate the impact of the subsidy on the probability of receiving welfare benefits and on unemployment status up to 72 months after programme start. We supplemented our estimation results with the results of a qualitative survey which provided insights into the diversity of entrepreneurial activities and the recipients' motivation for starting a business. Our results are encouraging. We found a stable positive effect on the outcomes of interest. We further found that foreign nationals and the long‐term unemployed benefit the most. We conclude that the subsidy allows some people to successfully start a business who would not have otherwise done so and who would have been less successful by choosing the alternative of continued job search.     

Feasible estimation in generalized structured models

This article introduces a feasible estimation method for a large class of semi and nonparametric models. We present the family of generalized structured models which we wish to estimate. After highlighting the main idea of the theoretical smooth backfitting estimators, we introduce a general estimation procedure. We consider modifications and practical issues, and discuss inference, cross validation, and asymptotic theory applying the theoretical framework of Mammen and Nielsen (Biometrika 90: 551–566, 2003). An extensive simulation study shows excellent performance of our method. Furthermore, real data applications from environmetrics and biometrics demonstrate its usefulness.     

Differences Between Household Income from Surveys and Registers and How These Affect the Poverty Headcount: Evidence from the Austrian SILC

We take advantage of the fact that for the Austrian SILC 2008–2011, two data sources are available in parallel for the same households: register-based and survey-based income data. Thus, we aim to explain which households tend to under- or over-report their household income by estimating multinomial logit and OLS models with covariates referring to the interview situation, employment status and socio-demographic household characteristics. Furthermore, we analyze source-specific differences in the distribution of household income and how these differences affect aggregate poverty indicators based on household income. The analysis reveals an increase in the cross-sectional poverty rates for 2008–2011 and the longitudinal poverty rate if register data rather than survey data are used. These changes in the poverty rate are mainly driven by differences in employment income rather than sampling weights and other income components. Regression results show a pattern of mean-reverting errors when comparing household income between the two data sources. Furthermore, differences between data sources for both under-reporting and over-reporting slightly decrease with the number of panel waves in which a household participated. Among the other variables analyzed that are related to the interview situation (mode, proxy, interview month), only the number of proxy interviews was (weakly) positively correlated with the difference between data sources, although this outcome was not robust over different model specifications.     

Digital document and interpretation: re-thinking “text” and scholarship in electronic settings

The contribution starts from outlining the evolution of the scholarly production flow from the print based paradigm to the digital age and in this context it explores the opposition of digital versus analog representation modes. It then develops on the triple paradigm shift caused by genuine digital publishing and its specific consequences for the social sciences and humanities (SSH) which in turn results in re-constituting basic scholarly notions such as ‘text’ and ‘document’. The paper concludes with discussing the specific value that could be added in systematically using digital text resources as a basis for scholarly work and also states some of the necessary conditions for such a ‘digital turn’ to be successful in the SSH.