Selection of influential variables in ordinal data with preponderance of zeros

Presence of excess zero in ordinal data is pervasive in areas like medical and social sciences. Unfortunately, analysis of such kind of data has so far hardly been looked into, perhaps for the reason that the underlying model that fits such data, is not a generalized linear model. Obviously some methodological developments and intensive computations are required. The current investigation is concerned with the selection of variables in such models. In many occasions where the number of predictors is quite large and some of them are not useful, the maximum likelihood approach is not the automatic choice. As, apart from the messy calculations involved, this approach fails to provide efficient estimates of the underlying parameters. The proposed penalized approach includes ?₁ penalty (LASSO) and the mixture of ?₁ and ?₂ penalties (elastic net). We propose a coordinate descent algorithm to fit a wide class of ordinal regression models and select useful variables appearing in both the ordinal regression and the logistic regression based mixing component. A rigorous discussion on the selection of predictors has been made through a simulation study. The proposed method is illustrated by analyzing the severity of driver injury from Michigan upper peninsula road accidents.     

Some characterizations of distributions by regression models for ordinal response data

Summary Let a random variableX be classified intok classes. By doing so, a new random variable is obtained, measured on ordinal scale. If this variable is a response variable in certain regression models for ordinal response data, the distribution ofX is characterized by the models. In this paper, characterizations of the distribution ofX by the proportional odds model and the proportional hazards model are given.     

Analysis of a mixed model for ordinal data by iterative re-weighted REML

An estimation procedure will be presented for a class of threshold models for ordinal data. These models may include both fixed and random effects with associated components of variance on an underlying scale. The residual error distribution on the underlying scale may be rendered greater flexibility by introducing additional shape parameters, e.g. a kurtosis parameter or parameters to model heterogeneous residual variances as a function of factors and covariates. The estimation procedure is an extension of an iterative re-weighted restricted maximum likelihood procedure, originally developed for generalized linear mixed models. This procedure will be illustrated with a practical problem involving damage to potato tubers and with data from animal breeding and medical research from the literature.     

A partial order toolbox for building synthetic indicators of sustainability with ordinal data

This paper designs and implements a statistical toolbox for the construction of synthetic indicators of sustainability and rankings of multidimensional systems of ordinal variables. The toolbox employs results from partial order theory to provide a purely ordinal way to perform the classical steps of indicator construction (i.e., with neither quantification nor aggregation of the input variables) and thus fills a gap in the statistical literature. “Ordinal” non-aggregative procedures are developed to score statistical units and assess the reliability of the final rankings as well as account for exogenous information relating to variable importance. The toolbox is introduced and shown in action through the real example of Sustainable Development Goal 16 (Peace, Justice, and Strong Institutions), specifically as it relates to Political Pluralism and Participation in North and South American countries.     

Robust analysis of longitudinal data with nonignorable missing responses

We encounter missing data in many longitudinal studies. When the missing data are nonignorable, it is important to analyze the data by incorporating the missing data mechanism into the observed data likelihood function. The classical maximum likelihood (ML) method for analyzing longitudinal missing data has been extensively studied in the literature. However, it is well-known that the ordinary ML estimators are sensitive to extreme observations or outliers in the data. In this paper, we propose and explore a robust method, which is developed in the framework of the ML method, and is useful for downweighting any influential observations in the data when estimating the model parameters. We study the empirical properties of the robust estimators in small simulations. We also illustrate the robust method using incomplete longitudinal data on CD4 counts from clinical trials of HIV-infected patients.     

培育适合中国特色的企业家队伍——“企业教练发展论坛”四人谈

由解放日报报业集团、经济日报经济研究中心、上海交通大学安泰管理学院和加拿大汇才人力技术有限公司共同主办的"二十一世纪企业教练发展论坛"日前在沪举行.     

Cluster Analysis for mixed data: An application to credit risk evaluation

Credit risk is one of the main risks faced by a bank to provide financial products and services to clients. To evaluate the financial performance of clients, several scoring methodologies have been proposed, which are based mostly on quantitative indicators. This paper highlights the relevance of both quantitative and qualitative features of applicants and proposes a new methodology based on mixed data clustering techniques. Indeed, cluster analysis may prove particularly useful in the estimation of credit risk. Traditionally, clustering concentrates only on quantitative or qualitative data at a time; however, since credit applicants are characterized by mixed personal features, a cluster analysis specific for mixed data can lead to discover particularly informative patterns, estimating the risk associated with credit granting.     

Using data envelopment analysis to measure the efficiency of non-profit organizations: A critical evaluation

This study examines the potential effects of variable set expansion and data variations upon the efficiency scores generated using the Data Envelopment Analysis (DEA) model. It was found that variable set expansion (either through disaggregation of existing variables or addition of new factors) should produce an upward trend in efficiency scores. In addition, ample opportunity exists for ‘decision-making units’ to increase their efficiency scores through manipulation of reported data. In real-world applications of DEA, these problems must be resolved as much as possible (e.g. increased audit of data) in order to improve DEA's practical usefulness and reliability.     

Production frontier estimation to measure efficiency: A critical evaluation in light of data envelopment analysis

The ranking and measurement of efficiency of decision-making units by two methods—data envelopment analysis and frontier production function—may not always lead to identical results. In this framework we attempt here a critical evaluation of the frontier production function theory in terms of theoretical and empirical implications. It is shown that under certain conditions the two approaches to effciency measurement may lead to identical results.     

Interns’ perceptions of HR practices and their influence on fit and intentions to join

Abstract

This study examines interns’ perceptions of HR practices and their influence on person–organization fit, person–job fit, and intentions to join the organization. The hypotheses are developed with signaling theory and attraction, selection, attrition (ASA) theory. Utilizing a time lagged study design of 71 interns employed in a United States based service firm, we found positive perceptions of HR practices rated early in the internship were related to increased person–organization fit, person–job fit, and intentions to join the organization at the end of the internship. Our findings also showed that person–organization fit and person–job fit fully mediated the relationship between HR practices and intentions to join the organization.     

Performance evaluation using data envelopment analysis in the presence of time lags

Data Envelopment Analysis (DEA) is a methodology that computes efficiency values for decision making units (DMU) in a given period by comparing the outputs with the inputs. In many applications, inputs and outputs of DMUs are monitored over time. There might be a time lag between the consumption of inputs and the production of outputs. We develop an approach that aims to capture the time lag between the outputs and the inputs in assigning the efficiency values to DMUs. We propose using weight restrictions in conjunction with the model. Our computational results on randomly generated problems demonstrate that the developed approach works well under a large variety of experimental conditions. We also apply our approach on a real data set to evaluate research institutions.     

A recommendation for applied researchers to substantiate the claim that ordinal variables are the product of underlying bivariate normal distributions

A simulation study was carried out to study the behaviour of the polychoric correlation coefficient in data not compliant with the assumption of underlying continuous variables. Such data can produce relatively high estimated polychoric correlations (in the order of .62). Applied researchers are prone to accept these artefacts as input for elaborate modelling (e.g., structural equation models) and inferences about reality justified by sheer magnitude of the correlations. In order to prevent this questionable research practice, it is recommended that in applications of the polychoric correlation coefficient, data is tested with goodness-of-fit of the BND, that such statistic is reported in published applications, and that the polychoric correlation is not applied when the test is significant.     

Efficiency evaluation of China's high-tech industry with a multi-activity network data envelopment analysis approach

This paper employs a Russell multi-activity network DEA model and divides the overall innovation process into the upstream Research and Development (R&D) process and the downstream commercialization process to appraise the innovation performance of China's high-tech industries from 2009 to 2013. This model can deal with the problems of intermediates, shared inputs and slack-based measure in a unified framework, and the result can provide policy makers with process-specific information on how to improve the innovation performance of China's high-tech industries. The main findings are presented as follows. First, the overall efficiency of China's high-tech industries still remains at a low level, which has its roots mainly in commercialization inefficiencies other than R&D inefficiencies. Second, for most provinces, their R&D efficiencies do not match up with their commercialization efficiencies. Finally, the innovative activities of China's high-tech industries should be driven by the market demand -oriented for the improvement of innovation efficiency.     

Robust prediction of domain compositions from uncertain data using isometric logratio transformations in a penalized multivariate Fay–Herriot model

Assessing regional population compositions is an important task in many research fields. Small area estimation with generalized linear mixed models marks a powerful tool for this purpose. However, the method has limitations in practice. When the data are subject to measurement errors, small area models produce inefficient or biased results since they cannot account for data uncertainty. This is particularly problematic for composition prediction, since generalized linear mixed models often rely on approximate likelihood inference. Obtained predictions are not reliable. We propose a robust multivariate Fay–Herriot model to solve these issues. It combines compositional data analysis with robust optimization theory. The nonlinear estimation of compositions is restated as a linear problem through isometric logratio transformations. Robust model parameter estimation is performed via penalized maximum likelihood. A robust best predictor is derived. Simulations are conducted to demonstrate the effectiveness of the approach. An application to alcohol consumption in Germany is provided.     

Linking of sample data to small areas

We consider methods for estimating the means of survey variables in domains of a finite population, where sample sizes are too small to obtain reliable direct estimates. We construct generalized compositions from the direct and traditional design-based synthetic estimators and propose the methodology for evaluating their coefficients. This methodology measures similarities among sample elements and estimates of the domain means. We propose the compositions for two cases of auxiliary information: domain-level characteristics are available; true means of auxiliary variables are available for the estimation domains, and unit-level auxiliary vectors are known for the sample elements. In the simulation study, we show where the generalized compositions improve the traditional synthetic and composite estimators.     

Integration of network data envelopment analysis and decision-making trial and evaluation laboratory for the performance evaluation of the financial holding companies in Taiwan

Data envelopment analysis (DEA) has been constantly used to measure the technical efficiency of decision-making units (DMUs). However, the major problem of traditional DEA methods is that they do not consider the possible intermediate effects. Recently, many papers have applied network DEA models to evaluate the efficiency scores. However, the linking activity of DMUs is still hard to be recognized. Hence, we employ DEMATEL to obtain the linking activity of DMUs. Our empirical research shows that the proposed method can soundly deal with the purpose of identifying the relationship between variables and derive the reasonable result in network DEA.     

The collapsibility theorem in log-linear analysis of categorical data: an application in program evaluation

The collapsibility theorem describes both the circumstances in which the effects of hierarchical models change when additional variables are introduced, as the circumstances in which the exclusion of certain variables and the analysis of specific marginal tables may lead to different conclusions. The partial association model is here considered as a specific example of three-dimensional log-linear analysis. Collapsibility is examined in an empirical study currently being performed in Catalonia with regard to program evaluation in penitentiary centers.     

On the usability of RAS in manpower forecasting: An evaluation based on Indian data

Projection of changes in the occupational structure is vital for ensuring accurate manpower forecasts. Yet this is seldom attempted in a rigorous way. One of the main difficulties which a manpower planner faces—especially in a developing economy—is the paucity of reliable time series data on the occupational pattern of employment. Given this constraint, it is maintained that it is futile to consider experimentation with sophisticated statistical forecasting techniques whose data requirements may be difficult to fulfil. The RAS, on the other hand, is a relatively simple technique which has been extensively used in adjusting matrices in diverse fields with only a given initial matrix and the row and column totals for another year. This is the first detailed application of the RAS to Indian occupational data. Our results confirm the potential usefulness of the RAS as a basis for approximating temporal changes in the occupational structure. Specifically, it is demonstrated here that as compared with the errors in occupational forecasts which would result from the assumption of an unchanging occupational structure, those associated with the RAS are markedly lower. The paper concludes by pointing out that, when detailed data become available, more flexible forms of the RAS which allow for the possibility of departures from biproportionality may be utilised.     

Doubly Robust Estimation of Causal Effects with Multivalued Treatments: An Application to the Returns to Schooling

This paper provides doubly robust estimators for treatment effect parameters which are defined in a multivalued treatment effect framework. We apply this method to the unique dataset of the 1970 British Cohort Study (BCS70) to estimate returns to various levels of schooling. The analysis is carried out for female and male samples separately to capture possible gender differences. Average returns are estimated for the entire population, as well as conditional on having a specific educational achievement. For males, relative to no qualification, we find an average return to O‐levels of 6.3%, to A‐levels of 7.9% and to higher education of 25.4%. The estimated average returns to O‐level and A‐level relative to no qualification are insignificant for females, whereas the return to higher education is 19.9%.Copyright © 2014 John Wiley & Sons, Ltd.