Complete Data Analysis Research Articles

Multiply-Imputed Synthetic Data: Advice to the Imputer

Abstract Several statistical agencies have started to use multiply-imputed synthetic microdata to create public-use data in major surveys. The purpose of doing this is to protect the confidentiality of respondents’ identities and sensitive attributes, while allowing standard complete-data analyses of microdata. A key challenge, faced by advocates of synthetic data, is demonstrating that valid statistical inferences can be obtained from such synthetic data for non-confidential questions. Large discrepancies between observed-data and synthetic-data analytic results for such questions may arise because of uncongeniality; that is, differences in the types of inputs available to the imputer, who has access to the actual data, and to the analyst, who has access only to the synthetic data. Here, we discuss a simple, but possibly canonical, example of uncongeniality when using multiple imputation to create synthetic data, which specifically addresses the choices made by the imputer. An initial, unanticipated but not surprising, conclusion is that non-confidential design information used to impute synthetic data should be released with the confidential synthetic data to allow users of synthetic data to avoid possible grossly conservative inferences.

Annealing effect on elastic, magnetic and magnetoelastic properties of CoFeB thin films on polymer substrate

We have studied the magnetomechanical properties of thin films deposited on polyimide substrate. Particularly, we studied the effect of annealing in the accessible temperature range (0–375 °C) allowed by the polymer substrate. Complementary techniques have been used to characterize the elastic, magnetic and magnetoelastic properties of annealed samples (Brillouin light scattering and tensile tests combined with in situ ferromagnetic resonance). Below 200 °C, no significant evolutions of the properties have been found excepted the magnetic damping parameter. In contrast, above 200 °C, strong variations are observed (+15% for Young’s modulus, for magnetostriction coefficient and +500% for the uniaxial magnetic anisotropy field). This is in correlation with the x-ray diffraction peak sharpening observed in the studied thin films. This is most probably due to exceeding of the glass transition temperature of the thin film. The complete data analysis allowed an estimation of the thermal stability of alloy thin films.

Weight Status

Background: Negative health behaviors such as consumption of excess calories, low intake of fruits and vegetables, sedentariness and weight gain are associated with entry into college. Purpose: To determine if weight status is associated with students’ receipt of health promotion (nutrition, physical activity and stress reduction) information, and students’ interest in receiving these types of information from their college or university. Methods: Data from the Spring 2011 ACHA-NCHA II dataset was used to complete secondary data analyses. Students (N=116,254) from 148 postsecondary institutions completed the Spring 2011 ACHA-NCHA II survey. Logistic regression was used to examine the effect of BMI category on receipt of, and interest in receiving, health promotion information. Results: Approximately 32% of respondents were overweight or obese. Students in the obese class III category were the least likely to receive health promotion information and least likely to be interested in receiving the information. Conclusion: Weight status based on BMI classification is a weak predictor of the dissemination of health promotion information. The largest gaps related to the dissemination appear to be among obese students. Future research is needed to determine factors contributing to the observed gaps and strategies should be developed to reach underserved groups.

Exploring injury severity measures and in-hospital mortality: A multi-hospital study in Kenya

IntroductionLow- and middle-income countries (LMICs) have a disproportionately high burden of injuries. Most injury severity measures were developed in high-income settings and there have been limited studies on their application and validity in low-resource settings. In this study, we compared the performance of seven injury severity measures: estimated Injury Severity Score (eISS), Glasgow Coma Score (GCS), Mechanism, GCS, Age, Pressure score (MGAP), GCS, Age, Pressure score (GAP), Revised Trauma Score (RTS), Trauma and Injury Severity Score (TRISS) and Kampala Trauma Score (KTS), in predicting in-hospital mortality in a multi-hospital cohort of adult patients in Kenya. MethodsThis study was performed using data from trauma registries implemented in four public hospitals in Kenya. Estimated ISS, MGAP, GAP, RTS, TRISS and KTS were computed according to algorithms described in the literature. All seven measures were compared for discrimination by computing area under curve (AUC) for the receiver operating characteristics (ROC), model fit information using Akaike information criterion (AIC), and model calibration curves. Sensitivity analysis was conducted to include all trauma patients during the study period who had missing information on any of the injury severity measure(s) through multiple imputations. ResultsA total of 16,548 patients were included in the study. Complete data analysis included 14,762 (90.2%) patients for the seven injury severity measures. TRISS (complete case AUC: 0.889, 95% CI: 0.866–0.907) and KTS (complete case AUC: 0.873, 95% CI: 0.852–0.892) demonstrated similarly better discrimination measured by AUC on in-hospital deaths overall in both complete case analysis and multiple imputations. Estimated ISS had lower AUC (0.764, 95% CI: 0.736–0.787) than some injury severity measures. Calibration plots showed eISS and RTS had lower calibration than models from other injury severity measures. ConclusionsThis multi-hospital study in Kenya found statistical significant higher performance of KTS and TRISS than other injury severity measures. The KTS, is however, an easier score to compute as compared to the TRISS and has stable good performance across several hospital settings and robust to missing values. It is therefore a practical and robust option for use in low-resource settings, and is applicable to settings similar to Kenya.

A Robust Missing Data-Recovering Technique for Mobility Data Mining

ABSTRACTBased on location information, users’ mobility profile building is the main task for making different useful systems such as early warning system, next destination and route prediction, tourist guide, mobile users’ behavior-aware applications, and potential friend recommendation. For mobility profile building, frequent trajectory patterns are required. The trajectory building is based on significant location extraction and the user’s actual movement prediction. Previous works have focused on significant places extraction without considering the change in GSM (global system for mobile communication) network and is based on complete data analysis. Since network operators change the GSM network periodically, there are possibilities of missing values and outliers. These missing values and outliers must be addressed to ensure actual mobility and for the efficient extraction of significant places, which are the basis for users’ trajectory building. In this paper, we propose a methodology to convert geo-coordinates into semantic tags and we also purposed a clustering methodology for recovering missing values and outlier detection. Experimental results prove the efficiency and effectiveness of the proposed scheme.

PyMICE: APython library for analysis of IntelliCage data

IntelliCage is an automated system for recording the behavior of a group of mice housed together. It produces rich, detailed behavioral data calling for new methods and software for their analysis. Here we present PyMICE, a free and open-source library for analysis of IntelliCage data in the Python programming language. We describe the design and demonstrate the use of the library through a series of examples. PyMICE provides easy and intuitive access to IntelliCage data, and thus facilitates the possibility of using numerous other Python scientific libraries to form a complete data analysis workflow.

Relationship between EASI and SCORAD severity assessments for atopic dermatitis

Capsule Summary The present study shows a complex relationship between EASI and SCORAD. The results provide support for inclusion of objective-SCORAD and/or SCORAD in addition to EASI in clinical trials.

Equity Return Modeling and Prediction Using Hybrid ARIMA-GARCH Model

In this paper, a hybrid ARIMA-GARCH model is proposed to model and predict the equity returns for three US benchmark indices: Dow Transportation, S&P 500 and VIX. Equity returns are univariate time series data sets, one of the methods to predict them is using the Auto-Regressive Integrated Moving Average (ARIMA) models. Despite the fact that the ARIMA models are powerful and flexible, they are not be able to handle the volatility and nonlinearity that are present in the time series data. However, the Generalized Autoregressive Conditional Heteroscedasticity (GARCH) models are designed to capture volatility clustering behavior in time series. In this paper, we provide motivations and descriptions of the hybrid ARIMA-GARCH model. A complete data analysis procedure that involves a series of hypothesis testings and a model fitting procedure using the Akaike Information Criterion (AIC) is provided in this paper as well. Simulation results of out of sample predictions are also provided in this paper as a reference.

Increasing research capacity with ICES Data & Analytic Services (DAS)

ABSTRACTBackgroundThe Institute for Clinical Evaluative Sciences (ICES) is a not-for-profit organization that conducts research to evaluate health care delivery and outcomes. Established in 1992, ICES houses a vast and secure array of linkable, coded health-related data on more than 13 million Ontarians, including health services data, health care provider data, registries and population-based health surveys. ICES has a reputation for generating strong evidence-based knowledge to inform policy and practice, however the use of the data was restricted to ICES’ purposes. In March 2014, ICES launched the Data & Analytic Services (DAS) platform with the primary objective of increasing access to its data to publicly-funded researchers, health care providers and administrators, policymakers and students.
 MethodDAS provides access to highly de-identified, risk-reduced datasets created from ICES’ data holdings; analytic support; and complete data analysis and report writing services. DAS also enables the importation of external data for linkage to ICES’ data holdings. Research objectives and methodology are led by the requestor and ICES analysts rely on their subject matter expertise to direct the deliverables.
 ResultsSince launch, over 200 requests from Canada, United States and United Kingdom have been adjudicated, of which 187 have been deemed feasible and eligible. Over the same period of time, ICES as an organization had over 700 active projects of which 348 were initiated, an increase in capacity of 26%. Though Toronto-based researchers represent the majority of the requests (62%), there have been requests from outside Ontario interested in comparing aspects of Ontario’s healthcare to their home province. Research topics have varied and include assessments of health care provision by sector, disease prevalence and treatment, and statistical methods. An unexpected outcome of increasing access has been the large interest from small physician groups and knowledge users who are not typically involved in research for academic purposes. Access to ICES’ data holdings provides an opportunity to examine a larger cohort of patients who share the same characteristics as their clinic patients or group. Furthermore, by enabling remote access to the data, DAS is able to leverage the capabilities of ICES’ data holdings and increase research capacity in a short period of time.
 ConclusionIn making one of the most comprehensively linked health administrative data repositories in the world widely available to the broader research and healthcare community, DAS engages investigators involved in front-line care, stimulates new avenues of research and fosters collaboration that was previously unachievable.

RNA-Seq Validation of RNAi Identifies Additional Gene Connectivity in Tribolium castaneum (Coleoptera: Tenebrionidae).

RNA interference (RNAi) is a functional genomics tool to correlate genotype and phenotype by delivering targeted, gene-specific, and complementary dsRNA into a host via injection, feeding, or other means in order to reduce gene expression. In the red flour beetle, Tribolium castaneum, RNAi has been successful via injected dsRNA at all life stages. Traditionally, successful transcript knockdown has been quantified by qPCR on a gene-by-gene basis, where only expression of the target gene and normalization genes are evaluated. In this study, RNA-Seq was used to quantify transcript expression in larvae injected with dsRNA for aspartate 1-decarboxylase (ADC), which gives a reliable phenotype of an adult with a black cuticle instead of the wild-type red-brown. ANOVA of control, mock-injected, and ADC-dsRNA injected larvae indicated that target gene expression was significantly (P = 0.002) reduced 4-fold, and the black phenotype was achieved in all adults injected with ADC-dsRNA as larvae. In a pairwise analysis, significant (P < 0.05) differential expression of other genes in ADC-injected larvae suggested connections between gene pathways. One gene, dopamine receptor 2, was increased in expression 227-fold (P = 0.025), presumably connected to previous data that showed a reduction in expression of ADC results in increased levels of dopamine. To evaluate the hypothesis that increased dopamine levels can affect mobility, T. castaneum adults injected with ADC-dsRNA as larvae were significantly impaired in movement tests compared to controls, similar to black mutants in Drosophila melanogaster. The data demonstrate that RNA-Seq can reveal gene connectivity and provide more complete data validation and analysis compared to qPCR.

Enhancing Interoperability and Capabilities of Earth Science Data using the Observations Data Model 2 (ODM2)

Earth Science researchers require access to integrated, cross-disciplinary data in order to answer critical research questions. Partially due to these science drivers, it is common for disciplinary data systems to expand from their original scope in order to accommodate collaborative research. The result is multiple disparate databases with overlapping but incompatible data. In order to enable more complete data integration and analysis, the Observations Data Model Version 2 (ODM2) was developed to be a general information model, with one of its major goals to integrate data collected by in situ sensors with those by ex-situ analyses of field specimens. Four use cases with different science drivers and disciplines have adopted ODM2 because of benefits to their users. The disciplines behind the four cases are diverse – hydrology, rock geochemistry, soil geochemistry, and biogeochemistry. For each case, we outline the benefits, challenges, and rationale for adopting ODM2. In each case, the decision to implement ODM2 was made to increase interoperability and expand data and metadata capabilities. One of the common benefits was the ability to use the flexible handling and comprehensive description of specimens and data collection sites in ODM2’s sampling feature concept. We also summarize best practices for implementing ODM2 based on the experience of these initial adopters. The descriptions here should help other potential adopters of ODM2 implement their own instances or to modify ODM2 to suit their needs.

Analysis of synonymous codon usage patterns in the edible fungus Volvariella volvacea.

In this study, trends in synonymous codons usage of Volvariella volvecea have been first examined by analysis of complete coding sequences and gene chip data. The results showed that GC content at three codon positions are obviously different and there were several factors shaping the codon usage of V. volvacea genes, including base composition. The comparison of codon usage among four edible fungi such as V. volvacea, Agaricus bisporus, Coprinopsis cinerea, and Pleurotus ostreatus indicated that the similar codon usage pattern was used among V. volvacea, A. bisporus and P. ostreatus, but there was significantly different codon usage pattern of C. cinerea. Two arrays of optimal codons were determined by effective number of codons (ENC) values and gene chip database separately, resulting that most of the ENC-predicted optimal codons were included in the array of gene chip resulted optimal codons. This study can provide useful information for codon usage pattern analysis and gene transformation of V. volvacea.

Analysis of complete, incomplete and total fusion data of 9Be +169Tm, 187Re, 209Bi reactions

We have analyzed the incomplete fusion (ICF), complete fusion (CF) and total fusion (TF) excitation functions for reactions induced by [Formula: see text] on [Formula: see text], [Formula: see text] and [Formula: see text] targets at near barrier energies using Wong’s formula in conjunction with the energy dependent Woods–Saxon potential. A phenomenological selection function is proposed to separate out the contribution of ICF and CF cross-sections in TF cross-section. The variation of relative contribution of ICF and CF in TF with respect to incident beam energy is very well reproduced through this approach.

Navigating Return to Work and Breastfeeding in a Hospital with a Comprehensive Employee Lactation Program.

The Surgeon General's Call to Action to Support Breastfeeding details the need for comprehensive employer lactation support programs. Our institution has an extensive employee lactation program, and our breastfeeding initiation and continuation rates are statistically significantly higher than state and national data, with more than 20% of our employees breastfeeding for more than 1 year. The objective of this research was complete secondary data analysis of qualitative data collected as part of a larger study on breastfeeding outcomes. In the larger study, 545 women who returned to work full or part time completed an online survey with the ability to provide free text qualitative data and feedback regarding their experiences with breastfeeding after return to work. Qualitative data were pulled from the online survey platform. The responses to these questions were analyzed using conventional content analysis by the research team (2 PhD-prepared nurse researchers trained and experienced in qualitative methodologies and 1 research assistant) in order to complete a thematic analysis of the survey data. Analysis of the data yielded 5 major themes: (1) positive reflections, (2) nonsupportive environment/work culture, (3) supportive environment/work culture, (4) accessibility of resources, and (5) internal barriers. The themes that emerged from this research clearly indicate that even in a hospital with an extensive employee lactation program, women have varied experiences-some more positive than others. Returning to work while breastfeeding requires time and commitment of the mother, and a supportive employee lactation program may ease that transition of return to work.

An alternative empirical likelihood method in missing response problems and causal inference.

Missing responses are common problems in medical, social, and economic studies. When responses are missing at random, a complete case data analysis may result in biases. A popular debias method is inverse probability weighting proposed by Horvitz and Thompson. To improve efficiency, Robins et al. proposed an augmented inverse probability weighting method. The augmented inverse probability weighting estimator has a double-robustness property and achieves the semiparametric efficiency lower bound when the regression model and propensity score model are both correctly specified. In this paper, we introduce an empirical likelihood-based estimator as an alternative to Qin and Zhang (2007). Our proposed estimator is also doubly robust and locally efficient. Simulation results show that the proposed estimator has better performance when the propensity score is correctly modeled. Moreover, the proposed method can be applied in the estimation of average treatment effect in observational causal inferences. Finally, we apply our method to an observational study of smoking, using data from the Cardiovascular Outcomes in Renal Atherosclerotic Lesions clinical trial. Copyright © 2016 John Wiley & Sons, Ltd.

Escape from the Laboratory: Ethnographic Methods in the Study of Elder and Family Court Mediation

The randomized control trial and pre/post research designs are commonly used in applied research and provide common standards for mediation evaluation research. These approaches have many benefits, particularly for evaluating whether mediation as an experimental intervention works or not.Scholars and practitioners, however, want to know not only whether mediation can work as expected but also how it works in a range of real‐world contexts over time. In these contexts, ideal experimental conditions are less likely to occur. Challenges include such circumstances as the following: the number of cases suitable for statistical comparison is insufficient; researchers lack control over how mediation is implemented; researchers lack clear, objective variables to measure; and the variability of mediation outcomes when studied over time makes it difficult to draw conclusions about them.My research has involved each of these challenges, and I have used ethnographic research as a way to evaluate mediation in these contexts. In this article, I explain ethnographic methods and present two studies as examples of mediation evaluation research that began with a standard program evaluation design, and then incorporated ethnography to allow more complete data collection and analysis. My purpose here is not to argue that experimental methods in mediation evaluation research should be displaced but rather to demonstrate how ethnographic methods can be used when the conditions necessary for standard evaluation cannot be met. The two studies used as examples are from an elder mediation study in Ghana and a family court mediation study in the United States.

Genomic Flatlining in the Endangered Island Fox

Genetic studies of rare and endangered species often focus on defining and preserving genetically distinct populations, especially those having unique adaptations [1, 2]. Much less attention is directed at understanding the landscape of deleterious variation, an insidious consequence of geographic isolation and the inefficiency of natural selection to eliminate harmful variants in small populations [3-5]. With population sizes of many vertebrates decreasing and isolation increasing through habitat fragmentation and loss, understanding the extent and nature of deleterious variation in small populations is essential for predicting and enhancing population persistence. The Channel Island fox (Urocyon littoralis) is a dwarfed species that inhabits six of California's Channel Islands andis derived from the mainland gray fox (U.cinereoargenteus). These isolated island populations have persisted for thousands of years at extremely small population sizes [6, 7] and, consequently, are a model for testing ideas about the accumulation of deleterious variation in small populations under natural conditions. Analysis of complete genome sequence data from island foxes shows a dramatic decrease in genome-wide variation and a sharp increase in the homozygosity of deleterious variants. The San Nicolas Island population has a near absence of variation, demonstrating a unique genetic flatlining that is punctuated by heterozygosity hotspots, enriched for olfactory receptor genes and other genes with high levels of ancestral variation. These findings question the generality of the small-population paradigm that maintains substantial genetic variation is necessary for short- and long-term persistence.

PAA: an R/bioconductor package for biomarker discovery with protein microarrays.

Summary: The R/Bioconductor package Protein Array Analyzer (PAA) facilitates a flexible analysis of protein microarrays for biomarker discovery (esp., ProtoArrays). It provides a complete data analysis workflow including preprocessing and quality control, uni- and multivariate feature selection as well as several different plots and results tables to outline and evaluate the analysis results. As a main feature, PAA’s multivariate feature selection methods are based on recursive feature elimination (e.g. SVM-recursive feature elimination, SVM-RFE) with stability ensuring strategies such as ensemble feature selection. This enables PAA to detect stable and reliable biomarker candidate panels.Availability and implementation: PAA is freely available (BSD 3-clause license) from http://www.bioconductor.org/packages/PAA/.Contact: michael.turewicz@rub.de or martin.eisenacher@rub.de

Radiometric Resolution Analysis and a Simulation Model

Total power radiometer has a simple configuration and the best theoretical resolution. Gain fluctuations and calibration errors, however, can induce severe errors in the solved scene brightness temperature. To estimate the overall radiometer performance we present a numerical simulation tool that can be used to determine the radiometric resolution. Our model considers three main components that degrade the radiometric resolution: thermal noise, 1/f noise and calibration errors. These error sources have long been known to exist, but comprehensive models able to account all these effects quantitatively and accurately in a practical manner have been missing. We have developed a radiometer simulation model that is able to produce radiometer signals that incorporate realistic radiometer effects that show up as noise and other errors in the radiometer video signal. Our simulation tool integrates the fundamental radiometer theories numerically and allows the investigation of different calibration schemes and receiver topologies. The model can be used as a guide for design and optimization as well as for verification of the radiometer performance. Moreover, it can be extended to a much larger and more complex radiometer systems allowing better system level performance estimation and optimization with minimal bread-board implementations. The model mimics real radiometer video data and thus the complete data analysis pipeline can be developed and verified before the real video data is available. In this paper, the model has been applied to a total power radiometer operating in the 52 GHz frequency range.

Complete wavefield seismic structural, stratigraphic, geomorphic and elastic interpretation in the offshore Browse Basin

The first quantitative interpretation project in the Browse Basin that used complete wavefields from dual-sensor streamer data to characterise a variety of overburden and reservoir seismic structural, sedimentary and geomorphic features is discussed in this extended abstract. Having established that complete wavefield pre-stack angle data are amplitude versus angle (AVA) compliant, the authors analyse the elastic response from thin gas sands in a location where well control predicts a weak gas response in terms of the Vp/Vs ratio and negligible response in terms of the acoustic impedance. Various seismic expressions in the near-surface are better resolved using complete wavefield imaging, including deltaic deposits, potentially prospective patch reefs and intersecting channel systems, and shallow hazards associated with historic drilling failures appear to be resolvable with greater confidence. On the larger scale, the overall data interpretability is substantially enhanced in texture and character by broadband imaging of primary reflections, and then by subsequent reservoir characterisation. Initial (deeper) interpretation focused around mapping of the prospective Early Jurassic Plover Formation and the Early Cretaceous Upper Vulcan Formation, in which evidence of channelisation can be demonstrated on conventional migration volumes and reinforced using complete wavefield migration. Ultra-low frequency signal penetration to the deeper Triassic level has been limited historically, but the first complete seismic wavefield data analysis presented demonstrates that a combination of volume attributes provide substantially improved geologic information from the seafloor to the deepest targets.