Nonparametric density deconvolution by weighted kernel estimators

Nonparametric density estimation in the presence of measurement error is considered. The usual kernel deconvolution estimator seeks to account for the contamination in the data by employing a modified kernel. In this paper a new approach based on a weighted kernel density estimator is proposed. Theoretical motivation is provided by the existence of a weight vector that perfectly counteracts the bias in density estimation without generating an excessive increase in variance. In practice a data driven method of weight selection is required. Our strategy is to minimize the discrepancy between a standard kernel estimate from the contaminated data on the one hand, and the convolution of the weighted deconvolution estimate with the measurement error density on the other hand. We consider a direct implementation of this approach, in which the weights are optimized subject to sum and non-negativity constraints, and a regularized version in which the objective function includes a ridge-type penalty. Numerical tests suggest that the weighted kernel estimation can lead to tangible improvements in performance over the usual kernel deconvolution estimator. Furthermore, weighted kernel estimates are free from the problem of negative estimation in the tails that can occur when using modified kernels. The weighted kernel approach generalizes to the case of multivariate deconvolution density estimation in a very straightforward manner.     

Monitoring the complexities: Nuclear power and public opinion

Interaction between organizations and stakeholders today takes place on virtual and physical ‘issue arenas’. This study examined opinions on nuclear power and asked who are the players discussing nuclear power in Finland? Through content analysis, surveys and interviews, the study concluded that politicians, power companies and regulators were the ones with voice, as NGOs and citizens were hardly heard. The paper suggests the future role of PR practitioners to be to find the right issue arenas and facilitate corporate voice and reputation on those arenas.     

Response of sensitive behaviors to frequent measurement

We study the influence of frequent survey measurement on behavior. Widespread access to the Internet has made important breakthroughs in frequent measurement possible—potentially revolutionizing social science measurement of processes that change quickly over time. One key concern about using such frequent measurement is that it may influence the behavior being studied. We investigate this possibility using both a population-based experiment with random assignment to participation in a weekly journal for twelve months (versus no journal) and a large-scale, population-based, journal-keeping study with weekly measurement for 30 months. Results reveal few of the measured behaviors are correlated with assignment to frequent measurement. Theoretical reasoning regarding the likely behavioral response to frequent measurement correctly predicts domains most vulnerable to this possibility. Overall, however, we found little evidence of behavioral response to frequent measurement.     

Generalized empirical likelihood inference in partially linear model for longitudinal data with missing response variables and error-prone covariates

In this article, we consider statistical inference for longitudinal partial linear models when the response variable is sometimes missing with missingness probability depending on the covariate that is measured with error. A generalized empirical likelihood (GEL) method is proposed by combining correction attenuation and quadratic inference functions. The method that takes into consideration the correlation within groups is used to estimate the regression coefficients. Furthermore, residual-adjusted empirical likelihood (EL) is employed for estimating the baseline function so that undersmoothing is avoided. The empirical log-likelihood ratios are proven to be asymptotically Chi-squared, and the corresponding confidence regions for the parameters of interest are then constructed. Compared with methods based on NAs, the GEL does not require consistent estimators for the asymptotic variance and bias. The numerical study is conducted to compare the performance of the EL and the normal approximation-based method, and a real example is analysed.     

Examining the interplay of an organization's prior reputation,CEO's visibility,and immediate response to a crisis

In one of only a few crisis communication research studies taking a relational approach, examining the effects of a company's prior reputation in publics’ responses to a given crisis situation, Lyons and Cameron (2004) found that both reputation and response profoundly affected publics’ attitude and behavioral intentions toward an organization involved in a crisis situation, using hypothetical, fictitious organizations and crises. Using actual organizations and crises, our research team designed a 2 (reputation: good vs. bad) × 2 (crisis response: apologetic vs. defensive) × 2 (CEO visibility in immediate crisis response: visible vs. invisible) within-subjects experiment (N = 102) to examine the variances in stakeholders’ attitudes and behavioral intentions toward a company after being exposed to online video that delivered a corporate crisis response. Findings were counter-intuitive: a defensive response to a crisis is as acceptable to crisis stakeholders as an apologetic response if the CEO is visible (or audible) in the response and if the pre-crisis company-stakeholder reputation is positive. Good reputation, defensive crisis response and CEO visibility in immediate response to a crisis resulted in the best stakeholder attitudes and purchase intentions.     

Nominal comparability is not enough: (In-)equivalence of construct validity of cross-national measures of educational attainment in the European Social Survey

Educational attainment is a core social background variable covered in each and every survey of individuals. Since educational institutions and qualifications are difficult to compare across countries, cross-national surveys pose a particular challenge to the measurement of educational attainment. This study performs a comparative construct validation of a number of cross-national measures of education using the European Social Survey. The measures comprise two versions of the International Standard Classification of Education (ISCED), the education scheme developed in the project ‘Comparative Analysis of Social Mobility in Industrial Nations’ (CASMIN) and hypothetical as well as actual years of education. The first ISCED measure corresponds to the well-known main ISCED levels. The second one, the European Survey Version of ISCED (ES-ISCED) developed for this study, represents an effort to reflect different types of education within levels of education by considering ISCED sub-dimensions, most importantly ‘programme orientation’. Using linear regression models, it is shown how much explanatory power educational attainment loses when different cross-national variables are used, as compared to country-specific educational attainment variables (CSEVs), and how these losses vary across measures and countries. The dependent variable used for the construct validation is social status as measured by the International Socio-Economic Index (ISEI). Results suggest that harmonisation always entails some loss of explanatory power for at least a few countries. However, there are clear performance differences between the comparable measures in terms of both the average amount of losses as well as the distribution of losses across countries. The use of actual years of education as well as the levels-only ISCED strongly attenuates the education-social status association on average, but also to very different degrees across countries. CASMIN and ES-ISCED fare considerably better: they show the lowest losses of explanatory power and the lowest variation of losses across countries. Hypothetical years of education lie in between. Some practical implications are then proposed, e.g. on how to implement cross-national measures of educational attainment in international surveys.     

Revisiting the gap between stylized and diary estimates of market work time

Previous studies find that workers of longer work weeks report more market hours in stylized measure than in time diary, while those of shorter work weeks report fewer hours. Using data from the American Time Use Survey (ATUS) 2003-2007, this paper attempts to explain this dilemma. I argue that market work time is composed of activity time, committed time, and constrained time. Thus, part of the difference found between the two estimates is due to the calculation of the diary estimate. I also argue that the wording of the stylized question is consequential. Some stylized questions (e.g., How many hours per week do you usually work?) might capture the mode, instead of the mean, of work time. Thus, when the distribution of work time is skewed, the mismatch between the two estimates is expected. A sequence-based identification for diary estimate and a non-parametric adjustment for stylized estimate are proposed and empirically examined. The result indicates that both methods significantly reduce the observed gap between the two estimates. I discuss the implications for future time use survey design in the concluding section.     

Nonparametric estimation of welfare changes

In this article, I propose a simple nonparametric method of estimating welfare measures with no structural assumptions on the demand system. The method is illustrated for both the single good and the multiple good case, using simulated data. Standard errors of the welfare changes are derived using standard bootstrap techniques. The compensating variation and deadweight loss resulting from a housing price increase is evaluated using Canadian mortgage data. The GAUSS coding, which is very short, is included.     

An Empirical Comparison of Estimating Methods for the Ramberg–Osgood EIV Model

The standard tensile test is one of the most frequent tools performed for the evaluation of mechanical properties of metals. An empirical model proposed by Ramberg and Osgood fits the tensile test data using a nonlinear model for the strain in terms of the stress. It is an Error-In-Variables (EIV) model because of the uncertainty affecting both strain and stress measurement instruments. The SIMEX, a simulation-based method for the estimation of model parameters, is powerful in order to reduce bias due to the measurement error in EIV models. The plan of this article is the following. In Sec. 2, we introduce the Ramberg–Osgood model and another reparametrization according to different assumptions on the independent variable. In Sec. 3, there is a summary of SIMEX method for the case at hand. Section 4 is a comparison between SIMEX and others estimating methods in order to highlight the peculiarities of the different approaches. In the last section, there are some concluding remarks.     

浅析高职院校信息化建设问题及对策

十二五期间，国家要求大力加强信息化建设，作为后起之秀，高职院校借此发展东风，迅速兴起信息化建设热潮。但在信息化建设中认识不到位、重复建设、效率低下、资源浪费严重等问题层出不穷。本文结合高职院校信息化实际，从四个方面分析了目前高职院校信息化建设中存在的问题，并提出切实改进意见，以期促进高职院校信息化建设高效、扎实稳步推进。